repo
stringlengths
1
152
file
stringlengths
14
221
code
stringlengths
501
25k
file_length
int64
501
25k
avg_line_length
float64
20
99.5
max_line_length
int64
21
134
extension_type
stringclasses
2 values
XNNPACK
XNNPACK-master/src/x64-transposec/gen/x64-transposec-4x4-multi-switch-avx.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/avx.c.in // Generator: tools/xngen // // Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <immintrin.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> #include <xnnpack/unaligned.h> void xnn_x64_transposec_ukernel__4x4_multi_switch_avx( const uint64_t* input, uint64_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x64_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(double)); assert(input_stride >= block_width * sizeof(double)); const size_t tile_height = 4; const size_t tile_width = 4; const size_t tile_hbytes = tile_height * sizeof(double); const size_t tile_wbytes = tile_width * sizeof(double); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t input_offset = tile_height * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(double); const double* i0 = (const double*) input; const double* i1 = (const double*) ((uintptr_t) i0 + input_stride); const double* i2 = (const double*) ((uintptr_t) i1 + input_stride); const double* i3 = (const double*) ((uintptr_t) i2 + input_stride); double* o = (double*) output; const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 3); const size_t oN_stride = rem * output_stride; __m256i vmask = _mm256_loadu_si256((const __m256i*) ((uintptr_t) &params->avx.mask_table[rem ^ 3])); size_t bh = block_height; for (; bh >= 4; bh -= 4) { const __m256d v2_0 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_offset); const __m256d v2_1 = _mm256_maskload_pd(i1, vmask); i1 = (double*) ((uintptr_t) i1 + input_offset); const __m256d v2_2 = _mm256_maskload_pd(i2, vmask); i2 = (double*) ((uintptr_t) i2 + input_offset); const __m256d v2_3 = _mm256_maskload_pd(i3, vmask); i3 = (double*) ((uintptr_t) i3 + input_offset); const __m256d v1_0 = _mm256_unpacklo_pd(v2_0, v2_1); const __m256d v1_1 = _mm256_unpackhi_pd(v2_0, v2_1); const __m256d v1_2 = _mm256_unpacklo_pd(v2_2, v2_3); const __m256d v1_3 = _mm256_unpackhi_pd(v2_2, v2_3); double* oN = (double*) ((uintptr_t) o + oN_stride); switch (rem) { default: XNN_UNREACHABLE; case 3: { const __m256d v0_3 = _mm256_permute2f128_pd(v1_1, v1_3, 0x31); _mm256_storeu_pd(oN, v0_3); oN = (double*) ((uintptr_t) oN + minus_output_stride); } case 2: { const __m256d v0_2 = _mm256_permute2f128_pd(v1_0, v1_2, 0x31); _mm256_storeu_pd(oN, v0_2); oN = (double*) ((uintptr_t) oN + minus_output_stride); } case 1: { const __m256d v0_1 = _mm256_insertf128_pd(v1_1, _mm256_castpd256_pd128(v1_3), 1); _mm256_storeu_pd( oN, v0_1); } case 0: { const __m256d v0_0 = _mm256_insertf128_pd(v1_0, _mm256_castpd256_pd128(v1_2), 1); _mm256_storeu_pd(o, v0_0); o = (double*) ((uintptr_t) o + tile_hbytes); } } } if (bh != 0) { const __m256d v2_0 = _mm256_maskload_pd(i0, vmask); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const __m256d v2_1 = _mm256_maskload_pd(i1, vmask); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i0; } const __m256d v2_2 = _mm256_maskload_pd(i2, vmask); const __m256d v2_3 = _mm256_undefined_pd(); const __m256d v1_0 = _mm256_unpacklo_pd(v2_0, v2_1); const __m256d v1_1 = _mm256_unpackhi_pd(v2_0, v2_1); const __m256d v1_2 = _mm256_unpacklo_pd(v2_2, v2_3); const __m256d v1_3 = _mm256_unpackhi_pd(v2_2, v2_3); __m256d v0_0 = _mm256_insertf128_pd(v1_0, _mm256_castpd256_pd128(v1_2), 1); __m256d v0_2 = _mm256_permute2f128_pd(v1_0, v1_2, 0x31); __m256d v0_1 = _mm256_insertf128_pd(v1_1, _mm256_castpd256_pd128(v1_3), 1); __m256d v0_3 = _mm256_permute2f128_pd(v1_1, v1_3, 0x31); __m128d v0_0_lo = _mm256_castpd256_pd128(v0_0); __m128d v0_1_lo = _mm256_castpd256_pd128(v0_1); __m128d v0_2_lo = _mm256_castpd256_pd128(v0_2); __m128d v0_3_lo = _mm256_castpd256_pd128(v0_3); if (bh & 2) { double* oN = (double*) ((uintptr_t) o + oN_stride); switch (rem) { case 3: _mm_storeu_pd(oN, v0_3_lo); v0_3_lo = _mm256_extractf128_pd(v0_3, 1); oN = (double*) ((uintptr_t) oN + minus_output_stride); case 2: _mm_storeu_pd(oN, v0_2_lo); v0_2_lo = _mm256_extractf128_pd(v0_2, 1); oN = (double*) ((uintptr_t) oN + minus_output_stride); case 1: _mm_storeu_pd(oN, v0_1_lo); v0_1_lo = _mm256_extractf128_pd(v0_1, 1); case 0: _mm_storeu_pd(o, v0_0_lo); v0_0_lo = _mm256_extractf128_pd(v0_0, 1); break; default: XNN_UNREACHABLE; } o += 2; } if (bh & 1) { double* oN = (double*) ((uintptr_t) o + oN_stride); switch (rem) { case 3: _mm_storel_pd(oN, v0_3_lo); oN = (double*) ((uintptr_t) oN + minus_output_stride); case 2: _mm_storel_pd(oN, v0_2_lo); oN = (double*) ((uintptr_t) oN + minus_output_stride); case 1: _mm_storel_pd(oN, v0_1_lo); case 0: _mm_storel_pd(o, v0_0_lo); break; default: XNN_UNREACHABLE; } } } i0 = (const double*) ((uintptr_t) i0 + input_reset); i1 = (const double*) ((uintptr_t) i0 + input_stride); i2 = (const double*) ((uintptr_t) i1 + input_stride); i3 = (const double*) ((uintptr_t) i2 + input_stride); o = (double*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
6,345
35.471264
108
c
XNNPACK
XNNPACK-master/src/x64-transposec/gen/x64-transposec-4x4-reuse-mov-avx.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/avx.c.in // Generator: tools/xngen // // Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <immintrin.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> #include <xnnpack/unaligned.h> void xnn_x64_transposec_ukernel__4x4_reuse_mov_avx( const uint64_t* input, uint64_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x64_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(double)); assert(input_stride >= block_width * sizeof(double)); const size_t tile_height = 4; const size_t tile_width = 4; const size_t tile_hbytes = tile_height * sizeof(double); const size_t tile_wbytes = tile_width * sizeof(double); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(double) - tile_hbytes; const double* i0 = (const double*) input; double* o = (double*) ((uintptr_t) output - tile_hbytes); const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 3); const size_t oN_stride = rem * output_stride; const size_t oN_offset = oN_stride + tile_hbytes; __m256i vmask = _mm256_loadu_si256((const __m256i*) ((uintptr_t) &params->avx.mask_table[rem ^ 3])); size_t bh = block_height; for (; bh >= 4; bh -= 4) { const __m256d v2_0 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v2_1 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v2_2 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v2_3 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v1_0 = _mm256_unpacklo_pd(v2_0, v2_1); const __m256d v1_1 = _mm256_unpackhi_pd(v2_0, v2_1); const __m256d v1_2 = _mm256_unpacklo_pd(v2_2, v2_3); const __m256d v1_3 = _mm256_unpackhi_pd(v2_2, v2_3); const __m256d v0_0 = _mm256_insertf128_pd(v1_0, _mm256_castpd256_pd128(v1_2), 1); const __m256d v0_2 = _mm256_permute2f128_pd(v1_0, v1_2, 0x31); const __m256d v0_1 = _mm256_insertf128_pd(v1_1, _mm256_castpd256_pd128(v1_3), 1); const __m256d v0_3 = _mm256_permute2f128_pd(v1_1, v1_3, 0x31); o = (double*) ((uintptr_t) o + oN_offset); _mm256_storeu_pd(o, v0_3); double *oN = (double*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } _mm256_storeu_pd(o, v0_2); oN = (double*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } _mm256_storeu_pd(o, v0_1); oN = (double*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } _mm256_storeu_pd(o, v0_0); } o = (double*) ((uintptr_t) o + tile_hbytes); if (bh != 0) { const __m256d v2_0 = _mm256_maskload_pd(i0, vmask); const double *i1 = (const double*) ((uintptr_t) i0 + input_stride); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const __m256d v2_1 = _mm256_maskload_pd(i1, vmask); const double *i2 = (const double*) ((uintptr_t) i1 + input_stride); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i1; } const __m256d v2_2 = _mm256_maskload_pd(i2, vmask); const __m256d v2_3 = _mm256_undefined_pd(); const __m256d v1_0 = _mm256_unpacklo_pd(v2_0, v2_1); const __m256d v1_1 = _mm256_unpackhi_pd(v2_0, v2_1); const __m256d v1_2 = _mm256_unpacklo_pd(v2_2, v2_3); const __m256d v1_3 = _mm256_unpackhi_pd(v2_2, v2_3); __m256d v0_0 = _mm256_insertf128_pd(v1_0, _mm256_castpd256_pd128(v1_2), 1); __m256d v0_2 = _mm256_permute2f128_pd(v1_0, v1_2, 0x31); __m256d v0_1 = _mm256_insertf128_pd(v1_1, _mm256_castpd256_pd128(v1_3), 1); __m256d v0_3 = _mm256_permute2f128_pd(v1_1, v1_3, 0x31); __m128d v0_0_lo = _mm256_castpd256_pd128(v0_0); __m128d v0_1_lo = _mm256_castpd256_pd128(v0_1); __m128d v0_2_lo = _mm256_castpd256_pd128(v0_2); __m128d v0_3_lo = _mm256_castpd256_pd128(v0_3); if (bh & 2) { o = (double*) ((uintptr_t) o + oN_stride); _mm_storeu_pd(o, v0_3_lo); v0_3_lo = _mm256_extractf128_pd(v0_3, 1); double *oN = (double*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } _mm_storeu_pd(o, v0_2_lo); v0_2_lo = _mm256_extractf128_pd(v0_2, 1); oN = (double*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } _mm_storeu_pd(o, v0_1_lo); v0_1_lo = _mm256_extractf128_pd(v0_1, 1); oN = (double*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } _mm_storeu_pd(o, v0_0_lo); v0_0_lo = _mm256_extractf128_pd(v0_0, 1); o += 2; } if (bh & 1) { o = (double*) ((uintptr_t) o + oN_stride); _mm_storel_pd(o, v0_3_lo); double *oN = (double*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } _mm_storel_pd(o, v0_2_lo); oN = (double*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } _mm_storel_pd(o, v0_1_lo); oN = (double*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } _mm_storel_pd(o, v0_0_lo); } } i0 = (const double*) ((uintptr_t) i0 + input_reset); o = (double*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
6,337
36.282353
122
c
XNNPACK
XNNPACK-master/src/x64-transposec/gen/x64-transposec-4x4-reuse-multi-avx.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/avx.c.in // Generator: tools/xngen // // Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <immintrin.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> #include <xnnpack/unaligned.h> void xnn_x64_transposec_ukernel__4x4_reuse_multi_avx( const uint64_t* input, uint64_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x64_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(double)); assert(input_stride >= block_width * sizeof(double)); const size_t tile_height = 4; const size_t tile_width = 4; const size_t tile_hbytes = tile_height * sizeof(double); const size_t tile_wbytes = tile_width * sizeof(double); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(double); const double* i0 = (const double*) input; double* o0 = (double*) output; do { double* o1 = (double*) (block_width < 2 ? o0 : (double*) ((uintptr_t) o0 + output_stride)); double* o2 = (double*) (block_width <= 2 ? o0 : (double*) ((uintptr_t) o1 + output_stride)); double* o3 = (double*) (block_width < 4 ? o0 : (double*) ((uintptr_t) o2 + output_stride)); const size_t rem = min(block_width - 1, 3); __m256i vmask = _mm256_loadu_si256((const __m256i*) ((uintptr_t) &params->avx.mask_table[rem ^ 3])); size_t bh = block_height; for (; bh >= 4; bh -= 4) { const __m256d v2_0 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v2_1 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v2_2 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v2_3 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v1_0 = _mm256_unpacklo_pd(v2_0, v2_1); const __m256d v1_1 = _mm256_unpackhi_pd(v2_0, v2_1); const __m256d v1_2 = _mm256_unpacklo_pd(v2_2, v2_3); const __m256d v1_3 = _mm256_unpackhi_pd(v2_2, v2_3); const __m256d v0_0 = _mm256_insertf128_pd(v1_0, _mm256_castpd256_pd128(v1_2), 1); const __m256d v0_2 = _mm256_permute2f128_pd(v1_0, v1_2, 0x31); const __m256d v0_1 = _mm256_insertf128_pd(v1_1, _mm256_castpd256_pd128(v1_3), 1); const __m256d v0_3 = _mm256_permute2f128_pd(v1_1, v1_3, 0x31); _mm256_storeu_pd(o3, v0_3); o3 = (double*) ((uintptr_t) o3 + tile_hbytes); _mm256_storeu_pd(o2, v0_2); o2 = (double*) ((uintptr_t) o2 + tile_hbytes); _mm256_storeu_pd(o1, v0_1); o1 = (double*) ((uintptr_t) o1 + tile_hbytes); _mm256_storeu_pd(o0, v0_0); o0 = (double*) ((uintptr_t) o0 + tile_hbytes); } if (bh != 0) { const __m256d v2_0 = _mm256_maskload_pd(i0, vmask); const double *i1 = (const double*) ((uintptr_t) i0 + input_stride); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const __m256d v2_1 = _mm256_maskload_pd(i1, vmask); const double *i2 = (const double*) ((uintptr_t) i1 + input_stride); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i1; } const __m256d v2_2 = _mm256_maskload_pd(i2, vmask); const __m256d v2_3 = _mm256_undefined_pd(); const __m256d v1_0 = _mm256_unpacklo_pd(v2_0, v2_1); const __m256d v1_1 = _mm256_unpackhi_pd(v2_0, v2_1); const __m256d v1_2 = _mm256_unpacklo_pd(v2_2, v2_3); const __m256d v1_3 = _mm256_unpackhi_pd(v2_2, v2_3); __m256d v0_0 = _mm256_insertf128_pd(v1_0, _mm256_castpd256_pd128(v1_2), 1); __m256d v0_2 = _mm256_permute2f128_pd(v1_0, v1_2, 0x31); __m256d v0_1 = _mm256_insertf128_pd(v1_1, _mm256_castpd256_pd128(v1_3), 1); __m256d v0_3 = _mm256_permute2f128_pd(v1_1, v1_3, 0x31); __m128d v0_0_lo = _mm256_castpd256_pd128(v0_0); __m128d v0_1_lo = _mm256_castpd256_pd128(v0_1); __m128d v0_2_lo = _mm256_castpd256_pd128(v0_2); __m128d v0_3_lo = _mm256_castpd256_pd128(v0_3); if (bh & 2) { _mm_storeu_pd(o3, v0_3_lo); v0_3_lo = _mm256_extractf128_pd(v0_3, 1); o3 += 2; _mm_storeu_pd(o2, v0_2_lo); v0_2_lo = _mm256_extractf128_pd(v0_2, 1); o2 += 2; _mm_storeu_pd(o1, v0_1_lo); v0_1_lo = _mm256_extractf128_pd(v0_1, 1); o1 += 2; _mm_storeu_pd(o0, v0_0_lo); v0_0_lo = _mm256_extractf128_pd(v0_0, 1); o0 += 2; } if (bh & 1) { _mm_storel_pd(o3, v0_3_lo); _mm_storel_pd(o2, v0_2_lo); _mm_storel_pd(o1, v0_1_lo); _mm_storel_pd(o0, v0_0_lo); } } i0 = (const double*) ((uintptr_t) i0 + input_reset); o0 = (double*) ((uintptr_t) o0 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
5,263
37.423358
108
c
XNNPACK
XNNPACK-master/src/x64-transposec/gen/x64-transposec-4x4-reuse-switch-avx.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/avx.c.in // Generator: tools/xngen // // Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <immintrin.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> #include <xnnpack/unaligned.h> void xnn_x64_transposec_ukernel__4x4_reuse_switch_avx( const uint64_t* input, uint64_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x64_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(double)); assert(input_stride >= block_width * sizeof(double)); const size_t tile_height = 4; const size_t tile_width = 4; const size_t tile_hbytes = tile_height * sizeof(double); const size_t tile_wbytes = tile_width * sizeof(double); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(double); const double* i0 = (const double*) input; double* o = (double*) output; const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 3); const size_t oN_stride = rem * output_stride; __m256i vmask = _mm256_loadu_si256((const __m256i*) ((uintptr_t) &params->avx.mask_table[rem ^ 3])); size_t bh = block_height; for (; bh >= 4; bh -= 4) { const __m256d v2_0 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v2_1 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v2_2 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v2_3 = _mm256_maskload_pd(i0, vmask); i0 = (double*) ((uintptr_t) i0 + input_stride); const __m256d v1_0 = _mm256_unpacklo_pd(v2_0, v2_1); const __m256d v1_1 = _mm256_unpackhi_pd(v2_0, v2_1); const __m256d v1_2 = _mm256_unpacklo_pd(v2_2, v2_3); const __m256d v1_3 = _mm256_unpackhi_pd(v2_2, v2_3); double* oN = (double*) ((uintptr_t) o + oN_stride); switch (rem) { default: XNN_UNREACHABLE; case 3: { const __m256d v0_3 = _mm256_permute2f128_pd(v1_1, v1_3, 0x31); _mm256_storeu_pd(oN, v0_3); oN = (double*) ((uintptr_t) oN + minus_output_stride); } case 2: { const __m256d v0_2 = _mm256_permute2f128_pd(v1_0, v1_2, 0x31); _mm256_storeu_pd(oN, v0_2); oN = (double*) ((uintptr_t) oN + minus_output_stride); } case 1: { const __m256d v0_1 = _mm256_insertf128_pd(v1_1, _mm256_castpd256_pd128(v1_3), 1); _mm256_storeu_pd( oN, v0_1); } case 0: { const __m256d v0_0 = _mm256_insertf128_pd(v1_0, _mm256_castpd256_pd128(v1_2), 1); _mm256_storeu_pd(o, v0_0); o = (double*) ((uintptr_t) o + tile_hbytes); } } } if (bh != 0) { const __m256d v2_0 = _mm256_maskload_pd(i0, vmask); const double *i1 = (const double*) ((uintptr_t) i0 + input_stride); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const __m256d v2_1 = _mm256_maskload_pd(i1, vmask); const double *i2 = (const double*) ((uintptr_t) i1 + input_stride); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i1; } const __m256d v2_2 = _mm256_maskload_pd(i2, vmask); const __m256d v2_3 = _mm256_undefined_pd(); const __m256d v1_0 = _mm256_unpacklo_pd(v2_0, v2_1); const __m256d v1_1 = _mm256_unpackhi_pd(v2_0, v2_1); const __m256d v1_2 = _mm256_unpacklo_pd(v2_2, v2_3); const __m256d v1_3 = _mm256_unpackhi_pd(v2_2, v2_3); __m256d v0_0 = _mm256_insertf128_pd(v1_0, _mm256_castpd256_pd128(v1_2), 1); __m256d v0_2 = _mm256_permute2f128_pd(v1_0, v1_2, 0x31); __m256d v0_1 = _mm256_insertf128_pd(v1_1, _mm256_castpd256_pd128(v1_3), 1); __m256d v0_3 = _mm256_permute2f128_pd(v1_1, v1_3, 0x31); __m128d v0_0_lo = _mm256_castpd256_pd128(v0_0); __m128d v0_1_lo = _mm256_castpd256_pd128(v0_1); __m128d v0_2_lo = _mm256_castpd256_pd128(v0_2); __m128d v0_3_lo = _mm256_castpd256_pd128(v0_3); if (bh & 2) { double* oN = (double*) ((uintptr_t) o + oN_stride); switch (rem) { case 3: _mm_storeu_pd(oN, v0_3_lo); v0_3_lo = _mm256_extractf128_pd(v0_3, 1); oN = (double*) ((uintptr_t) oN + minus_output_stride); case 2: _mm_storeu_pd(oN, v0_2_lo); v0_2_lo = _mm256_extractf128_pd(v0_2, 1); oN = (double*) ((uintptr_t) oN + minus_output_stride); case 1: _mm_storeu_pd(oN, v0_1_lo); v0_1_lo = _mm256_extractf128_pd(v0_1, 1); case 0: _mm_storeu_pd(o, v0_0_lo); v0_0_lo = _mm256_extractf128_pd(v0_0, 1); break; default: XNN_UNREACHABLE; } o += 2; } if (bh & 1) { double* oN = (double*) ((uintptr_t) o + oN_stride); switch (rem) { case 3: _mm_storel_pd(oN, v0_3_lo); oN = (double*) ((uintptr_t) oN + minus_output_stride); case 2: _mm_storel_pd(oN, v0_2_lo); oN = (double*) ((uintptr_t) oN + minus_output_stride); case 1: _mm_storel_pd(oN, v0_1_lo); case 0: _mm_storel_pd(o, v0_0_lo); break; default: XNN_UNREACHABLE; } } } i0 = (const double*) ((uintptr_t) i0 + input_reset); o = (double*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
6,051
34.810651
108
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-aarch64-neon-tbx128x4-x16.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/neon-tbx128x4.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <arm_neon.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__aarch64_neon_tbx128x4_x16( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const uint8x16x4_t vtable0123 = vld1q_u8_x4(table); const uint8x16x4_t vtable4567 = vld1q_u8_x4(table + 64); const uint8x16x4_t vtable89AB = vld1q_u8_x4(table + 128); const uint8x16x4_t vtableCDEF = vld1q_u8_x4(table + 192); const uint8x16_t voffset = vmovq_n_u8(64); for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { uint8x16_t vx = vld1q_u8(input); input += 16; uint8x16_t vy = vqtbl4q_u8(vtable0123, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable4567, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable89AB, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtableCDEF, vx); vst1q_u8(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { uint8x16_t vx = vld1q_u8(input); uint8x16_t vy = vqtbl4q_u8(vtable0123, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable4567, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable89AB, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtableCDEF, vx); uint8x8_t vy_lo = vget_low_u8(vy); if (batch & (8 * sizeof(uint8_t))) { vst1_u8(output, vy_lo); output += 8; vy_lo = vget_high_u8(vy); } if (batch & (4 * sizeof(uint8_t))) { vst1_lane_u32((void*) output, vreinterpret_u32_u8(vy_lo), 0); output += 4; vy_lo = vext_u8(vy_lo, vy_lo, 4); } if (batch & (2 * sizeof(uint8_t))) { vst1_lane_u16((void*) output, vreinterpret_u16_u8(vy_lo), 0); output += 2; vy_lo = vext_u8(vy_lo, vy_lo, 2); } if (batch & (1 * sizeof(uint8_t))) { vst1_lane_u8(output, vy_lo, 0); } } }
2,354
27.373494
80
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-aarch64-neon-tbx128x4-x32.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/neon-tbx128x4.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <arm_neon.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__aarch64_neon_tbx128x4_x32( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const uint8x16x4_t vtable0123 = vld1q_u8_x4(table); const uint8x16x4_t vtable4567 = vld1q_u8_x4(table + 64); const uint8x16x4_t vtable89AB = vld1q_u8_x4(table + 128); const uint8x16x4_t vtableCDEF = vld1q_u8_x4(table + 192); const uint8x16_t voffset = vmovq_n_u8(64); for (; batch >= 32 * sizeof(uint8_t); batch -= 32 * sizeof(uint8_t)) { uint8x16_t vx0 = vld1q_u8(input); input += 16; uint8x16_t vx1 = vld1q_u8(input); input += 16; uint8x16_t vy0 = vqtbl4q_u8(vtable0123, vx0); vx0 = vsubq_u8(vx0, voffset); uint8x16_t vy1 = vqtbl4q_u8(vtable0123, vx1); vx1 = vsubq_u8(vx1, voffset); vy0 = vqtbx4q_u8(vy0, vtable4567, vx0); vx0 = vsubq_u8(vx0, voffset); vy1 = vqtbx4q_u8(vy1, vtable4567, vx1); vx1 = vsubq_u8(vx1, voffset); vy0 = vqtbx4q_u8(vy0, vtable89AB, vx0); vx0 = vsubq_u8(vx0, voffset); vy1 = vqtbx4q_u8(vy1, vtable89AB, vx1); vx1 = vsubq_u8(vx1, voffset); vy0 = vqtbx4q_u8(vy0, vtableCDEF, vx0); vy1 = vqtbx4q_u8(vy1, vtableCDEF, vx1); vst1q_u8(output, vy0); output += 16; vst1q_u8(output, vy1); output += 16; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { uint8x16_t vx = vld1q_u8(input); input += 16; uint8x16_t vy = vqtbl4q_u8(vtable0123, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable4567, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable89AB, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtableCDEF, vx); vst1q_u8(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { uint8x16_t vx = vld1q_u8(input); uint8x16_t vy = vqtbl4q_u8(vtable0123, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable4567, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable89AB, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtableCDEF, vx); uint8x8_t vy_lo = vget_low_u8(vy); if (batch & (8 * sizeof(uint8_t))) { vst1_u8(output, vy_lo); output += 8; vy_lo = vget_high_u8(vy); } if (batch & (4 * sizeof(uint8_t))) { vst1_lane_u32((void*) output, vreinterpret_u32_u8(vy_lo), 0); output += 4; vy_lo = vext_u8(vy_lo, vy_lo, 4); } if (batch & (2 * sizeof(uint8_t))) { vst1_lane_u16((void*) output, vreinterpret_u16_u8(vy_lo), 0); output += 2; vy_lo = vext_u8(vy_lo, vy_lo, 2); } if (batch & (1 * sizeof(uint8_t))) { vst1_lane_u8(output, vy_lo, 0); } } }
3,188
28.527778
80
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-aarch64-neon-tbx128x4-x48.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/neon-tbx128x4.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <arm_neon.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__aarch64_neon_tbx128x4_x48( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const uint8x16x4_t vtable0123 = vld1q_u8_x4(table); const uint8x16x4_t vtable4567 = vld1q_u8_x4(table + 64); const uint8x16x4_t vtable89AB = vld1q_u8_x4(table + 128); const uint8x16x4_t vtableCDEF = vld1q_u8_x4(table + 192); const uint8x16_t voffset = vmovq_n_u8(64); for (; batch >= 48 * sizeof(uint8_t); batch -= 48 * sizeof(uint8_t)) { uint8x16_t vx0 = vld1q_u8(input); input += 16; uint8x16_t vx1 = vld1q_u8(input); input += 16; uint8x16_t vx2 = vld1q_u8(input); input += 16; uint8x16_t vy0 = vqtbl4q_u8(vtable0123, vx0); vx0 = vsubq_u8(vx0, voffset); uint8x16_t vy1 = vqtbl4q_u8(vtable0123, vx1); vx1 = vsubq_u8(vx1, voffset); uint8x16_t vy2 = vqtbl4q_u8(vtable0123, vx2); vx2 = vsubq_u8(vx2, voffset); vy0 = vqtbx4q_u8(vy0, vtable4567, vx0); vx0 = vsubq_u8(vx0, voffset); vy1 = vqtbx4q_u8(vy1, vtable4567, vx1); vx1 = vsubq_u8(vx1, voffset); vy2 = vqtbx4q_u8(vy2, vtable4567, vx2); vx2 = vsubq_u8(vx2, voffset); vy0 = vqtbx4q_u8(vy0, vtable89AB, vx0); vx0 = vsubq_u8(vx0, voffset); vy1 = vqtbx4q_u8(vy1, vtable89AB, vx1); vx1 = vsubq_u8(vx1, voffset); vy2 = vqtbx4q_u8(vy2, vtable89AB, vx2); vx2 = vsubq_u8(vx2, voffset); vy0 = vqtbx4q_u8(vy0, vtableCDEF, vx0); vy1 = vqtbx4q_u8(vy1, vtableCDEF, vx1); vy2 = vqtbx4q_u8(vy2, vtableCDEF, vx2); vst1q_u8(output, vy0); output += 16; vst1q_u8(output, vy1); output += 16; vst1q_u8(output, vy2); output += 16; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { uint8x16_t vx = vld1q_u8(input); input += 16; uint8x16_t vy = vqtbl4q_u8(vtable0123, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable4567, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable89AB, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtableCDEF, vx); vst1q_u8(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { uint8x16_t vx = vld1q_u8(input); uint8x16_t vy = vqtbl4q_u8(vtable0123, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable4567, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable89AB, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtableCDEF, vx); uint8x8_t vy_lo = vget_low_u8(vy); if (batch & (8 * sizeof(uint8_t))) { vst1_u8(output, vy_lo); output += 8; vy_lo = vget_high_u8(vy); } if (batch & (4 * sizeof(uint8_t))) { vst1_lane_u32((void*) output, vreinterpret_u32_u8(vy_lo), 0); output += 4; vy_lo = vext_u8(vy_lo, vy_lo, 4); } if (batch & (2 * sizeof(uint8_t))) { vst1_lane_u16((void*) output, vreinterpret_u16_u8(vy_lo), 0); output += 2; vy_lo = vext_u8(vy_lo, vy_lo, 2); } if (batch & (1 * sizeof(uint8_t))) { vst1_lane_u8(output, vy_lo, 0); } } }
3,564
29.470085
80
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-aarch64-neon-tbx128x4-x64.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/neon-tbx128x4.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <arm_neon.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__aarch64_neon_tbx128x4_x64( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const uint8x16x4_t vtable0123 = vld1q_u8_x4(table); const uint8x16x4_t vtable4567 = vld1q_u8_x4(table + 64); const uint8x16x4_t vtable89AB = vld1q_u8_x4(table + 128); const uint8x16x4_t vtableCDEF = vld1q_u8_x4(table + 192); const uint8x16_t voffset = vmovq_n_u8(64); for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { uint8x16_t vx0 = vld1q_u8(input); input += 16; uint8x16_t vx1 = vld1q_u8(input); input += 16; uint8x16_t vx2 = vld1q_u8(input); input += 16; uint8x16_t vx3 = vld1q_u8(input); input += 16; uint8x16_t vy0 = vqtbl4q_u8(vtable0123, vx0); vx0 = vsubq_u8(vx0, voffset); uint8x16_t vy1 = vqtbl4q_u8(vtable0123, vx1); vx1 = vsubq_u8(vx1, voffset); uint8x16_t vy2 = vqtbl4q_u8(vtable0123, vx2); vx2 = vsubq_u8(vx2, voffset); uint8x16_t vy3 = vqtbl4q_u8(vtable0123, vx3); vx3 = vsubq_u8(vx3, voffset); vy0 = vqtbx4q_u8(vy0, vtable4567, vx0); vx0 = vsubq_u8(vx0, voffset); vy1 = vqtbx4q_u8(vy1, vtable4567, vx1); vx1 = vsubq_u8(vx1, voffset); vy2 = vqtbx4q_u8(vy2, vtable4567, vx2); vx2 = vsubq_u8(vx2, voffset); vy3 = vqtbx4q_u8(vy3, vtable4567, vx3); vx3 = vsubq_u8(vx3, voffset); vy0 = vqtbx4q_u8(vy0, vtable89AB, vx0); vx0 = vsubq_u8(vx0, voffset); vy1 = vqtbx4q_u8(vy1, vtable89AB, vx1); vx1 = vsubq_u8(vx1, voffset); vy2 = vqtbx4q_u8(vy2, vtable89AB, vx2); vx2 = vsubq_u8(vx2, voffset); vy3 = vqtbx4q_u8(vy3, vtable89AB, vx3); vx3 = vsubq_u8(vx3, voffset); vy0 = vqtbx4q_u8(vy0, vtableCDEF, vx0); vy1 = vqtbx4q_u8(vy1, vtableCDEF, vx1); vy2 = vqtbx4q_u8(vy2, vtableCDEF, vx2); vy3 = vqtbx4q_u8(vy3, vtableCDEF, vx3); vst1q_u8(output, vy0); output += 16; vst1q_u8(output, vy1); output += 16; vst1q_u8(output, vy2); output += 16; vst1q_u8(output, vy3); output += 16; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { uint8x16_t vx = vld1q_u8(input); input += 16; uint8x16_t vy = vqtbl4q_u8(vtable0123, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable4567, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable89AB, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtableCDEF, vx); vst1q_u8(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { uint8x16_t vx = vld1q_u8(input); uint8x16_t vy = vqtbl4q_u8(vtable0123, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable4567, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtable89AB, vx); vx = vsubq_u8(vx, voffset); vy = vqtbx4q_u8(vy, vtableCDEF, vx); uint8x8_t vy_lo = vget_low_u8(vy); if (batch & (8 * sizeof(uint8_t))) { vst1_u8(output, vy_lo); output += 8; vy_lo = vget_high_u8(vy); } if (batch & (4 * sizeof(uint8_t))) { vst1_lane_u32((void*) output, vreinterpret_u32_u8(vy_lo), 0); output += 4; vy_lo = vext_u8(vy_lo, vy_lo, 4); } if (batch & (2 * sizeof(uint8_t))) { vst1_lane_u16((void*) output, vreinterpret_u16_u8(vy_lo), 0); output += 2; vy_lo = vext_u8(vy_lo, vy_lo, 2); } if (batch & (1 * sizeof(uint8_t))) { vst1_lane_u8(output, vy_lo, 0); } } }
3,940
30.277778
80
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx-x16.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/ssse3.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/common.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> void xnn_x8_lut_ukernel__avx_x16( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m128i vt0 = _mm_load_si128((const __m128i*) table); const __m128i vt1 = _mm_load_si128((const __m128i*) (table + 16)); const __m128i vt2 = _mm_load_si128((const __m128i*) (table + 32)); const __m128i vt3 = _mm_load_si128((const __m128i*) (table + 48)); const __m128i vt4 = _mm_load_si128((const __m128i*) (table + 64)); const __m128i vt5 = _mm_load_si128((const __m128i*) (table + 80)); const __m128i vt6 = _mm_load_si128((const __m128i*) (table + 96)); const __m128i vt7 = _mm_load_si128((const __m128i*) (table + 112)); const __m128i vt8 = _mm_load_si128((const __m128i*) (table + 128)); const __m128i vt9 = _mm_load_si128((const __m128i*) (table + 144)); const __m128i vtA = _mm_load_si128((const __m128i*) (table + 160)); const __m128i vtB = _mm_load_si128((const __m128i*) (table + 176)); const __m128i vtC = _mm_load_si128((const __m128i*) (table + 192)); const __m128i vtD = _mm_load_si128((const __m128i*) (table + 208)); const __m128i vtE = _mm_load_si128((const __m128i*) (table + 224)); const __m128i vtF = _mm_load_si128((const __m128i*) (table + 240)); const __m128i vtable0 = vt0; const __m128i vtable1 = _mm_xor_si128(vt0, vt1); const __m128i vtable2 = _mm_xor_si128(vt1, vt2); const __m128i vtable3 = _mm_xor_si128(vt2, vt3); const __m128i vtable4 = _mm_xor_si128(vt3, vt4); const __m128i vtable5 = _mm_xor_si128(vt4, vt5); const __m128i vtable6 = _mm_xor_si128(vt5, vt6); const __m128i vtable7 = _mm_xor_si128(vt6, vt7); const __m128i vtable8 = _mm_xor_si128(_mm_xor_si128(vt7, vt8), vtable0); const __m128i vtable9 = _mm_xor_si128(_mm_xor_si128(vt8, vt9), vtable1); const __m128i vtableA = _mm_xor_si128(_mm_xor_si128(vt9, vtA), vtable2); const __m128i vtableB = _mm_xor_si128(_mm_xor_si128(vtA, vtB), vtable3); const __m128i vtableC = _mm_xor_si128(_mm_xor_si128(vtB, vtC), vtable4); const __m128i vtableD = _mm_xor_si128(_mm_xor_si128(vtC, vtD), vtable5); const __m128i vtableE = _mm_xor_si128(_mm_xor_si128(vtD, vtE), vtable6); const __m128i vtableF = _mm_xor_si128(_mm_xor_si128(vtE, vtF), vtable7); const __m128i voffset = _mm_set1_epi8(16); for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { _mm_storeu_si32(output, vy); vy = _mm_srli_epi64(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { _mm_storeu_si16(output, vy); vy = _mm_srli_epi32(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) _mm_extract_epi8(vy, 0); } } }
6,625
39.650307
74
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx-x32.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/ssse3.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/common.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> void xnn_x8_lut_ukernel__avx_x32( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m128i vt0 = _mm_load_si128((const __m128i*) table); const __m128i vt1 = _mm_load_si128((const __m128i*) (table + 16)); const __m128i vt2 = _mm_load_si128((const __m128i*) (table + 32)); const __m128i vt3 = _mm_load_si128((const __m128i*) (table + 48)); const __m128i vt4 = _mm_load_si128((const __m128i*) (table + 64)); const __m128i vt5 = _mm_load_si128((const __m128i*) (table + 80)); const __m128i vt6 = _mm_load_si128((const __m128i*) (table + 96)); const __m128i vt7 = _mm_load_si128((const __m128i*) (table + 112)); const __m128i vt8 = _mm_load_si128((const __m128i*) (table + 128)); const __m128i vt9 = _mm_load_si128((const __m128i*) (table + 144)); const __m128i vtA = _mm_load_si128((const __m128i*) (table + 160)); const __m128i vtB = _mm_load_si128((const __m128i*) (table + 176)); const __m128i vtC = _mm_load_si128((const __m128i*) (table + 192)); const __m128i vtD = _mm_load_si128((const __m128i*) (table + 208)); const __m128i vtE = _mm_load_si128((const __m128i*) (table + 224)); const __m128i vtF = _mm_load_si128((const __m128i*) (table + 240)); const __m128i vtable0 = vt0; const __m128i vtable1 = _mm_xor_si128(vt0, vt1); const __m128i vtable2 = _mm_xor_si128(vt1, vt2); const __m128i vtable3 = _mm_xor_si128(vt2, vt3); const __m128i vtable4 = _mm_xor_si128(vt3, vt4); const __m128i vtable5 = _mm_xor_si128(vt4, vt5); const __m128i vtable6 = _mm_xor_si128(vt5, vt6); const __m128i vtable7 = _mm_xor_si128(vt6, vt7); const __m128i vtable8 = _mm_xor_si128(_mm_xor_si128(vt7, vt8), vtable0); const __m128i vtable9 = _mm_xor_si128(_mm_xor_si128(vt8, vt9), vtable1); const __m128i vtableA = _mm_xor_si128(_mm_xor_si128(vt9, vtA), vtable2); const __m128i vtableB = _mm_xor_si128(_mm_xor_si128(vtA, vtB), vtable3); const __m128i vtableC = _mm_xor_si128(_mm_xor_si128(vtB, vtC), vtable4); const __m128i vtableD = _mm_xor_si128(_mm_xor_si128(vtC, vtD), vtable5); const __m128i vtableE = _mm_xor_si128(_mm_xor_si128(vtD, vtE), vtable6); const __m128i vtableF = _mm_xor_si128(_mm_xor_si128(vtE, vtF), vtable7); const __m128i voffset = _mm_set1_epi8(16); for (; batch >= 32 * sizeof(uint8_t); batch -= 32 * sizeof(uint8_t)) { __m128i vx0 = _mm_loadu_si128((const __m128i*) input); __m128i vx1 = _mm_loadu_si128((const __m128i*) (input + 16)); input += 32; __m128i vy0 = _mm_shuffle_epi8(vtable0, vx0); __m128i vy1 = _mm_shuffle_epi8(vtable0, vx1); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable1, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable1, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable2, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable2, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable3, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable3, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable4, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable4, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable5, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable5, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable6, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable6, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable7, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable7, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable8, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable8, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable9, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable9, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableA, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableA, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableB, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableB, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableC, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableC, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableD, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableD, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableE, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableE, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableF, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableF, vx1)); _mm_storeu_si128((__m128i*) output, vy0); _mm_storeu_si128((__m128i*) (output + 16), vy1); output += 32; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { _mm_storeu_si32(output, vy); vy = _mm_srli_epi64(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { _mm_storeu_si16(output, vy); vy = _mm_srli_epi32(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) _mm_extract_epi8(vy, 0); } } }
10,079
41.531646
74
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx-x48.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/ssse3.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/common.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> void xnn_x8_lut_ukernel__avx_x48( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m128i vt0 = _mm_load_si128((const __m128i*) table); const __m128i vt1 = _mm_load_si128((const __m128i*) (table + 16)); const __m128i vt2 = _mm_load_si128((const __m128i*) (table + 32)); const __m128i vt3 = _mm_load_si128((const __m128i*) (table + 48)); const __m128i vt4 = _mm_load_si128((const __m128i*) (table + 64)); const __m128i vt5 = _mm_load_si128((const __m128i*) (table + 80)); const __m128i vt6 = _mm_load_si128((const __m128i*) (table + 96)); const __m128i vt7 = _mm_load_si128((const __m128i*) (table + 112)); const __m128i vt8 = _mm_load_si128((const __m128i*) (table + 128)); const __m128i vt9 = _mm_load_si128((const __m128i*) (table + 144)); const __m128i vtA = _mm_load_si128((const __m128i*) (table + 160)); const __m128i vtB = _mm_load_si128((const __m128i*) (table + 176)); const __m128i vtC = _mm_load_si128((const __m128i*) (table + 192)); const __m128i vtD = _mm_load_si128((const __m128i*) (table + 208)); const __m128i vtE = _mm_load_si128((const __m128i*) (table + 224)); const __m128i vtF = _mm_load_si128((const __m128i*) (table + 240)); const __m128i vtable0 = vt0; const __m128i vtable1 = _mm_xor_si128(vt0, vt1); const __m128i vtable2 = _mm_xor_si128(vt1, vt2); const __m128i vtable3 = _mm_xor_si128(vt2, vt3); const __m128i vtable4 = _mm_xor_si128(vt3, vt4); const __m128i vtable5 = _mm_xor_si128(vt4, vt5); const __m128i vtable6 = _mm_xor_si128(vt5, vt6); const __m128i vtable7 = _mm_xor_si128(vt6, vt7); const __m128i vtable8 = _mm_xor_si128(_mm_xor_si128(vt7, vt8), vtable0); const __m128i vtable9 = _mm_xor_si128(_mm_xor_si128(vt8, vt9), vtable1); const __m128i vtableA = _mm_xor_si128(_mm_xor_si128(vt9, vtA), vtable2); const __m128i vtableB = _mm_xor_si128(_mm_xor_si128(vtA, vtB), vtable3); const __m128i vtableC = _mm_xor_si128(_mm_xor_si128(vtB, vtC), vtable4); const __m128i vtableD = _mm_xor_si128(_mm_xor_si128(vtC, vtD), vtable5); const __m128i vtableE = _mm_xor_si128(_mm_xor_si128(vtD, vtE), vtable6); const __m128i vtableF = _mm_xor_si128(_mm_xor_si128(vtE, vtF), vtable7); const __m128i voffset = _mm_set1_epi8(16); for (; batch >= 48 * sizeof(uint8_t); batch -= 48 * sizeof(uint8_t)) { __m128i vx0 = _mm_loadu_si128((const __m128i*) input); __m128i vx1 = _mm_loadu_si128((const __m128i*) (input + 16)); __m128i vx2 = _mm_loadu_si128((const __m128i*) (input + 32)); input += 48; __m128i vy0 = _mm_shuffle_epi8(vtable0, vx0); __m128i vy1 = _mm_shuffle_epi8(vtable0, vx1); __m128i vy2 = _mm_shuffle_epi8(vtable0, vx2); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable1, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable1, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable1, vx2)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable2, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable2, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable2, vx2)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable3, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable3, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable3, vx2)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable4, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable4, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable4, vx2)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable5, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable5, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable5, vx2)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable6, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable6, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable6, vx2)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable7, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable7, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable7, vx2)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable8, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable8, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable8, vx2)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable9, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable9, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable9, vx2)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableA, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableA, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableA, vx2)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableB, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableB, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableB, vx2)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableC, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableC, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableC, vx2)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableD, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableD, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableD, vx2)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableE, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableE, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableE, vx2)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableF, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableF, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableF, vx2)); _mm_storeu_si128((__m128i*) output, vy0); _mm_storeu_si128((__m128i*) (output + 16), vy1); _mm_storeu_si128((__m128i*) (output + 32), vy2); output += 48; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { _mm_storeu_si32(output, vy); vy = _mm_srli_epi64(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { _mm_storeu_si16(output, vy); vy = _mm_srli_epi32(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) _mm_extract_epi8(vy, 0); } } }
11,755
42.540741
74
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx-x64.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/ssse3.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/common.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> void xnn_x8_lut_ukernel__avx_x64( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m128i vt0 = _mm_load_si128((const __m128i*) table); const __m128i vt1 = _mm_load_si128((const __m128i*) (table + 16)); const __m128i vt2 = _mm_load_si128((const __m128i*) (table + 32)); const __m128i vt3 = _mm_load_si128((const __m128i*) (table + 48)); const __m128i vt4 = _mm_load_si128((const __m128i*) (table + 64)); const __m128i vt5 = _mm_load_si128((const __m128i*) (table + 80)); const __m128i vt6 = _mm_load_si128((const __m128i*) (table + 96)); const __m128i vt7 = _mm_load_si128((const __m128i*) (table + 112)); const __m128i vt8 = _mm_load_si128((const __m128i*) (table + 128)); const __m128i vt9 = _mm_load_si128((const __m128i*) (table + 144)); const __m128i vtA = _mm_load_si128((const __m128i*) (table + 160)); const __m128i vtB = _mm_load_si128((const __m128i*) (table + 176)); const __m128i vtC = _mm_load_si128((const __m128i*) (table + 192)); const __m128i vtD = _mm_load_si128((const __m128i*) (table + 208)); const __m128i vtE = _mm_load_si128((const __m128i*) (table + 224)); const __m128i vtF = _mm_load_si128((const __m128i*) (table + 240)); const __m128i vtable0 = vt0; const __m128i vtable1 = _mm_xor_si128(vt0, vt1); const __m128i vtable2 = _mm_xor_si128(vt1, vt2); const __m128i vtable3 = _mm_xor_si128(vt2, vt3); const __m128i vtable4 = _mm_xor_si128(vt3, vt4); const __m128i vtable5 = _mm_xor_si128(vt4, vt5); const __m128i vtable6 = _mm_xor_si128(vt5, vt6); const __m128i vtable7 = _mm_xor_si128(vt6, vt7); const __m128i vtable8 = _mm_xor_si128(_mm_xor_si128(vt7, vt8), vtable0); const __m128i vtable9 = _mm_xor_si128(_mm_xor_si128(vt8, vt9), vtable1); const __m128i vtableA = _mm_xor_si128(_mm_xor_si128(vt9, vtA), vtable2); const __m128i vtableB = _mm_xor_si128(_mm_xor_si128(vtA, vtB), vtable3); const __m128i vtableC = _mm_xor_si128(_mm_xor_si128(vtB, vtC), vtable4); const __m128i vtableD = _mm_xor_si128(_mm_xor_si128(vtC, vtD), vtable5); const __m128i vtableE = _mm_xor_si128(_mm_xor_si128(vtD, vtE), vtable6); const __m128i vtableF = _mm_xor_si128(_mm_xor_si128(vtE, vtF), vtable7); const __m128i voffset = _mm_set1_epi8(16); for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m128i vx0 = _mm_loadu_si128((const __m128i*) input); __m128i vx1 = _mm_loadu_si128((const __m128i*) (input + 16)); __m128i vx2 = _mm_loadu_si128((const __m128i*) (input + 32)); __m128i vx3 = _mm_loadu_si128((const __m128i*) (input + 48)); input += 64; __m128i vy0 = _mm_shuffle_epi8(vtable0, vx0); __m128i vy1 = _mm_shuffle_epi8(vtable0, vx1); __m128i vy2 = _mm_shuffle_epi8(vtable0, vx2); __m128i vy3 = _mm_shuffle_epi8(vtable0, vx3); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vx3 = _mm_sub_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable1, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable1, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable1, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtable1, vx3)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vx3 = _mm_sub_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable2, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable2, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable2, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtable2, vx3)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vx3 = _mm_sub_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable3, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable3, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable3, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtable3, vx3)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vx3 = _mm_sub_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable4, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable4, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable4, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtable4, vx3)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vx3 = _mm_sub_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable5, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable5, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable5, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtable5, vx3)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vx3 = _mm_sub_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable6, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable6, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable6, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtable6, vx3)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vx3 = _mm_sub_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable7, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable7, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable7, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtable7, vx3)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vx2 = _mm_sub_epi8(vx2, voffset); vx3 = _mm_sub_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable8, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable8, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable8, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtable8, vx3)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vx3 = _mm_subs_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable9, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable9, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtable9, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtable9, vx3)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vx3 = _mm_subs_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableA, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableA, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableA, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtableA, vx3)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vx3 = _mm_subs_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableB, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableB, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableB, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtableB, vx3)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vx3 = _mm_subs_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableC, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableC, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableC, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtableC, vx3)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vx3 = _mm_subs_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableD, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableD, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableD, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtableD, vx3)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vx3 = _mm_subs_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableE, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableE, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableE, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtableE, vx3)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vx2 = _mm_subs_epi8(vx2, voffset); vx3 = _mm_subs_epi8(vx3, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableF, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableF, vx1)); vy2 = _mm_xor_si128(vy2, _mm_shuffle_epi8(vtableF, vx2)); vy3 = _mm_xor_si128(vy3, _mm_shuffle_epi8(vtableF, vx3)); _mm_storeu_si128((__m128i*) output, vy0); _mm_storeu_si128((__m128i*) (output + 16), vy1); _mm_storeu_si128((__m128i*) (output + 32), vy2); _mm_storeu_si128((__m128i*) (output + 48), vy3); output += 64; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { _mm_storeu_si32(output, vy); vy = _mm_srli_epi64(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { _mm_storeu_si16(output, vy); vy = _mm_srli_epi32(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) _mm_extract_epi8(vy, 0); } } }
13,431
43.330033
74
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx2-x128.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx2.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx2_x128( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m256i vt0 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) table)); const __m256i vt1 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 16))); const __m256i vt2 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 32))); const __m256i vt3 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 48))); const __m256i vt4 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 64))); const __m256i vt5 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 80))); const __m256i vt6 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 96))); const __m256i vt7 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 112))); const __m256i vt8 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 128))); const __m256i vt9 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 144))); const __m256i vtA = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 160))); const __m256i vtB = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 176))); const __m256i vtC = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 192))); const __m256i vtD = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 208))); const __m256i vtE = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 224))); const __m256i vtF = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 240))); const __m256i vtable0 = vt0; const __m256i vtable1 = _mm256_xor_si256(vt0, vt1); const __m256i vtable2 = _mm256_xor_si256(vt1, vt2); const __m256i vtable3 = _mm256_xor_si256(vt2, vt3); const __m256i vtable4 = _mm256_xor_si256(vt3, vt4); const __m256i vtable5 = _mm256_xor_si256(vt4, vt5); const __m256i vtable6 = _mm256_xor_si256(vt5, vt6); const __m256i vtable7 = _mm256_xor_si256(vt6, vt7); const __m256i vtable8 = _mm256_xor_si256(_mm256_xor_si256(vt7, vt8), vtable0); const __m256i vtable9 = _mm256_xor_si256(_mm256_xor_si256(vt8, vt9), vtable1); const __m256i vtableA = _mm256_xor_si256(_mm256_xor_si256(vt9, vtA), vtable2); const __m256i vtableB = _mm256_xor_si256(_mm256_xor_si256(vtA, vtB), vtable3); const __m256i vtableC = _mm256_xor_si256(_mm256_xor_si256(vtB, vtC), vtable4); const __m256i vtableD = _mm256_xor_si256(_mm256_xor_si256(vtC, vtD), vtable5); const __m256i vtableE = _mm256_xor_si256(_mm256_xor_si256(vtD, vtE), vtable6); const __m256i vtableF = _mm256_xor_si256(_mm256_xor_si256(vtE, vtF), vtable7); const __m256i voffset = _mm256_set1_epi8(16); for (; batch >= 128 * sizeof(uint8_t); batch -= 128 * sizeof(uint8_t)) { __m256i vx0 = _mm256_loadu_si256((const __m256i*) input); __m256i vx1 = _mm256_loadu_si256((const __m256i*) (input + 32)); __m256i vx2 = _mm256_loadu_si256((const __m256i*) (input + 64)); __m256i vx3 = _mm256_loadu_si256((const __m256i*) (input + 96)); input += 128; __m256i vy0 = _mm256_shuffle_epi8(vtable0, vx0); __m256i vy1 = _mm256_shuffle_epi8(vtable0, vx1); __m256i vy2 = _mm256_shuffle_epi8(vtable0, vx2); __m256i vy3 = _mm256_shuffle_epi8(vtable0, vx3); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vx3 = _mm256_sub_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable1, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable1, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable1, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtable1, vx3)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vx3 = _mm256_sub_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable2, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable2, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable2, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtable2, vx3)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vx3 = _mm256_sub_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable3, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable3, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable3, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtable3, vx3)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vx3 = _mm256_sub_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable4, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable4, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable4, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtable4, vx3)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vx3 = _mm256_sub_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable5, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable5, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable5, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtable5, vx3)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vx3 = _mm256_sub_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable6, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable6, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable6, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtable6, vx3)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vx3 = _mm256_sub_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable7, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable7, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable7, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtable7, vx3)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vx3 = _mm256_sub_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable8, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable8, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable8, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtable8, vx3)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vx3 = _mm256_subs_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable9, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable9, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable9, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtable9, vx3)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vx3 = _mm256_subs_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableA, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableA, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableA, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtableA, vx3)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vx3 = _mm256_subs_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableB, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableB, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableB, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtableB, vx3)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vx3 = _mm256_subs_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableC, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableC, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableC, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtableC, vx3)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vx3 = _mm256_subs_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableD, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableD, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableD, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtableD, vx3)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vx3 = _mm256_subs_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableE, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableE, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableE, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtableE, vx3)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vx3 = _mm256_subs_epi8(vx3, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableF, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableF, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableF, vx2)); vy3 = _mm256_xor_si256(vy3, _mm256_shuffle_epi8(vtableF, vx3)); _mm256_storeu_si256((__m256i*) output, vy0); _mm256_storeu_si256((__m256i*) (output + 32), vy1); _mm256_storeu_si256((__m256i*) (output + 64), vy2); _mm256_storeu_si256((__m256i*) (output + 96), vy3); output += 128; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(_mm256_castsi256_si128(vtable0), vx); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable1), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable2), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable3), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable4), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable5), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable6), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable7), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable8), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable9), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableA), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableB), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableC), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableD), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableE), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableF), vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(_mm256_castsi256_si128(vtable0), vx); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable1), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable2), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable3), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable4), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable5), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable6), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable7), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable8), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable9), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableA), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableB), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableC), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableD), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableE), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableF), vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { _mm_storeu_si32(output, vy); vy = _mm_srli_epi64(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { _mm_storeu_si16(output, vy); vy = _mm_srli_epi32(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) _mm_extract_epi8(vy, 0); } } }
16,036
51.927393
98
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx2-x32.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx2.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx2_x32( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m256i vt0 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) table)); const __m256i vt1 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 16))); const __m256i vt2 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 32))); const __m256i vt3 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 48))); const __m256i vt4 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 64))); const __m256i vt5 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 80))); const __m256i vt6 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 96))); const __m256i vt7 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 112))); const __m256i vt8 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 128))); const __m256i vt9 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 144))); const __m256i vtA = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 160))); const __m256i vtB = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 176))); const __m256i vtC = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 192))); const __m256i vtD = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 208))); const __m256i vtE = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 224))); const __m256i vtF = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 240))); const __m256i vtable0 = vt0; const __m256i vtable1 = _mm256_xor_si256(vt0, vt1); const __m256i vtable2 = _mm256_xor_si256(vt1, vt2); const __m256i vtable3 = _mm256_xor_si256(vt2, vt3); const __m256i vtable4 = _mm256_xor_si256(vt3, vt4); const __m256i vtable5 = _mm256_xor_si256(vt4, vt5); const __m256i vtable6 = _mm256_xor_si256(vt5, vt6); const __m256i vtable7 = _mm256_xor_si256(vt6, vt7); const __m256i vtable8 = _mm256_xor_si256(_mm256_xor_si256(vt7, vt8), vtable0); const __m256i vtable9 = _mm256_xor_si256(_mm256_xor_si256(vt8, vt9), vtable1); const __m256i vtableA = _mm256_xor_si256(_mm256_xor_si256(vt9, vtA), vtable2); const __m256i vtableB = _mm256_xor_si256(_mm256_xor_si256(vtA, vtB), vtable3); const __m256i vtableC = _mm256_xor_si256(_mm256_xor_si256(vtB, vtC), vtable4); const __m256i vtableD = _mm256_xor_si256(_mm256_xor_si256(vtC, vtD), vtable5); const __m256i vtableE = _mm256_xor_si256(_mm256_xor_si256(vtD, vtE), vtable6); const __m256i vtableF = _mm256_xor_si256(_mm256_xor_si256(vtE, vtF), vtable7); const __m256i voffset = _mm256_set1_epi8(16); for (; batch >= 32 * sizeof(uint8_t); batch -= 32 * sizeof(uint8_t)) { __m256i vx0 = _mm256_loadu_si256((const __m256i*) input); input += 32; __m256i vy0 = _mm256_shuffle_epi8(vtable0, vx0); vx0 = _mm256_sub_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable1, vx0)); vx0 = _mm256_sub_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable2, vx0)); vx0 = _mm256_sub_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable3, vx0)); vx0 = _mm256_sub_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable4, vx0)); vx0 = _mm256_sub_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable5, vx0)); vx0 = _mm256_sub_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable6, vx0)); vx0 = _mm256_sub_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable7, vx0)); vx0 = _mm256_sub_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable8, vx0)); vx0 = _mm256_subs_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable9, vx0)); vx0 = _mm256_subs_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableA, vx0)); vx0 = _mm256_subs_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableB, vx0)); vx0 = _mm256_subs_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableC, vx0)); vx0 = _mm256_subs_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableD, vx0)); vx0 = _mm256_subs_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableE, vx0)); vx0 = _mm256_subs_epi8(vx0, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableF, vx0)); _mm256_storeu_si256((__m256i*) output, vy0); output += 32; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(_mm256_castsi256_si128(vtable0), vx); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable1), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable2), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable3), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable4), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable5), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable6), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable7), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable8), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable9), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableA), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableB), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableC), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableD), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableE), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableF), vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(_mm256_castsi256_si128(vtable0), vx); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable1), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable2), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable3), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable4), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable5), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable6), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable7), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable8), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable9), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableA), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableB), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableC), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableD), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableE), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableF), vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { _mm_storeu_si32(output, vy); vy = _mm_srli_epi64(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { _mm_storeu_si16(output, vy); vy = _mm_srli_epi32(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) _mm_extract_epi8(vy, 0); } } }
10,571
50.823529
98
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx2-x64.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx2.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx2_x64( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m256i vt0 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) table)); const __m256i vt1 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 16))); const __m256i vt2 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 32))); const __m256i vt3 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 48))); const __m256i vt4 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 64))); const __m256i vt5 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 80))); const __m256i vt6 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 96))); const __m256i vt7 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 112))); const __m256i vt8 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 128))); const __m256i vt9 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 144))); const __m256i vtA = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 160))); const __m256i vtB = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 176))); const __m256i vtC = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 192))); const __m256i vtD = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 208))); const __m256i vtE = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 224))); const __m256i vtF = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 240))); const __m256i vtable0 = vt0; const __m256i vtable1 = _mm256_xor_si256(vt0, vt1); const __m256i vtable2 = _mm256_xor_si256(vt1, vt2); const __m256i vtable3 = _mm256_xor_si256(vt2, vt3); const __m256i vtable4 = _mm256_xor_si256(vt3, vt4); const __m256i vtable5 = _mm256_xor_si256(vt4, vt5); const __m256i vtable6 = _mm256_xor_si256(vt5, vt6); const __m256i vtable7 = _mm256_xor_si256(vt6, vt7); const __m256i vtable8 = _mm256_xor_si256(_mm256_xor_si256(vt7, vt8), vtable0); const __m256i vtable9 = _mm256_xor_si256(_mm256_xor_si256(vt8, vt9), vtable1); const __m256i vtableA = _mm256_xor_si256(_mm256_xor_si256(vt9, vtA), vtable2); const __m256i vtableB = _mm256_xor_si256(_mm256_xor_si256(vtA, vtB), vtable3); const __m256i vtableC = _mm256_xor_si256(_mm256_xor_si256(vtB, vtC), vtable4); const __m256i vtableD = _mm256_xor_si256(_mm256_xor_si256(vtC, vtD), vtable5); const __m256i vtableE = _mm256_xor_si256(_mm256_xor_si256(vtD, vtE), vtable6); const __m256i vtableF = _mm256_xor_si256(_mm256_xor_si256(vtE, vtF), vtable7); const __m256i voffset = _mm256_set1_epi8(16); for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m256i vx0 = _mm256_loadu_si256((const __m256i*) input); __m256i vx1 = _mm256_loadu_si256((const __m256i*) (input + 32)); input += 64; __m256i vy0 = _mm256_shuffle_epi8(vtable0, vx0); __m256i vy1 = _mm256_shuffle_epi8(vtable0, vx1); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable1, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable1, vx1)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable2, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable2, vx1)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable3, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable3, vx1)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable4, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable4, vx1)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable5, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable5, vx1)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable6, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable6, vx1)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable7, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable7, vx1)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable8, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable8, vx1)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable9, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable9, vx1)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableA, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableA, vx1)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableB, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableB, vx1)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableC, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableC, vx1)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableD, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableD, vx1)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableE, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableE, vx1)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableF, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableF, vx1)); _mm256_storeu_si256((__m256i*) output, vy0); _mm256_storeu_si256((__m256i*) (output + 32), vy1); output += 64; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(_mm256_castsi256_si128(vtable0), vx); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable1), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable2), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable3), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable4), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable5), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable6), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable7), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable8), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable9), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableA), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableB), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableC), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableD), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableE), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableF), vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(_mm256_castsi256_si128(vtable0), vx); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable1), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable2), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable3), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable4), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable5), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable6), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable7), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable8), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable9), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableA), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableB), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableC), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableD), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableE), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableF), vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { _mm_storeu_si32(output, vy); vy = _mm_srli_epi64(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { _mm_storeu_si16(output, vy); vy = _mm_srli_epi32(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) _mm_extract_epi8(vy, 0); } } }
12,391
51.28692
98
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx2-x96.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx2.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx2_x96( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m256i vt0 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) table)); const __m256i vt1 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 16))); const __m256i vt2 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 32))); const __m256i vt3 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 48))); const __m256i vt4 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 64))); const __m256i vt5 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 80))); const __m256i vt6 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 96))); const __m256i vt7 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 112))); const __m256i vt8 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 128))); const __m256i vt9 = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 144))); const __m256i vtA = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 160))); const __m256i vtB = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 176))); const __m256i vtC = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 192))); const __m256i vtD = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 208))); const __m256i vtE = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 224))); const __m256i vtF = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) (table + 240))); const __m256i vtable0 = vt0; const __m256i vtable1 = _mm256_xor_si256(vt0, vt1); const __m256i vtable2 = _mm256_xor_si256(vt1, vt2); const __m256i vtable3 = _mm256_xor_si256(vt2, vt3); const __m256i vtable4 = _mm256_xor_si256(vt3, vt4); const __m256i vtable5 = _mm256_xor_si256(vt4, vt5); const __m256i vtable6 = _mm256_xor_si256(vt5, vt6); const __m256i vtable7 = _mm256_xor_si256(vt6, vt7); const __m256i vtable8 = _mm256_xor_si256(_mm256_xor_si256(vt7, vt8), vtable0); const __m256i vtable9 = _mm256_xor_si256(_mm256_xor_si256(vt8, vt9), vtable1); const __m256i vtableA = _mm256_xor_si256(_mm256_xor_si256(vt9, vtA), vtable2); const __m256i vtableB = _mm256_xor_si256(_mm256_xor_si256(vtA, vtB), vtable3); const __m256i vtableC = _mm256_xor_si256(_mm256_xor_si256(vtB, vtC), vtable4); const __m256i vtableD = _mm256_xor_si256(_mm256_xor_si256(vtC, vtD), vtable5); const __m256i vtableE = _mm256_xor_si256(_mm256_xor_si256(vtD, vtE), vtable6); const __m256i vtableF = _mm256_xor_si256(_mm256_xor_si256(vtE, vtF), vtable7); const __m256i voffset = _mm256_set1_epi8(16); for (; batch >= 96 * sizeof(uint8_t); batch -= 96 * sizeof(uint8_t)) { __m256i vx0 = _mm256_loadu_si256((const __m256i*) input); __m256i vx1 = _mm256_loadu_si256((const __m256i*) (input + 32)); __m256i vx2 = _mm256_loadu_si256((const __m256i*) (input + 64)); input += 96; __m256i vy0 = _mm256_shuffle_epi8(vtable0, vx0); __m256i vy1 = _mm256_shuffle_epi8(vtable0, vx1); __m256i vy2 = _mm256_shuffle_epi8(vtable0, vx2); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable1, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable1, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable1, vx2)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable2, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable2, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable2, vx2)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable3, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable3, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable3, vx2)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable4, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable4, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable4, vx2)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable5, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable5, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable5, vx2)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable6, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable6, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable6, vx2)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable7, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable7, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable7, vx2)); vx0 = _mm256_sub_epi8(vx0, voffset); vx1 = _mm256_sub_epi8(vx1, voffset); vx2 = _mm256_sub_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable8, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable8, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable8, vx2)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtable9, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtable9, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtable9, vx2)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableA, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableA, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableA, vx2)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableB, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableB, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableB, vx2)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableC, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableC, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableC, vx2)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableD, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableD, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableD, vx2)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableE, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableE, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableE, vx2)); vx0 = _mm256_subs_epi8(vx0, voffset); vx1 = _mm256_subs_epi8(vx1, voffset); vx2 = _mm256_subs_epi8(vx2, voffset); vy0 = _mm256_xor_si256(vy0, _mm256_shuffle_epi8(vtableF, vx0)); vy1 = _mm256_xor_si256(vy1, _mm256_shuffle_epi8(vtableF, vx1)); vy2 = _mm256_xor_si256(vy2, _mm256_shuffle_epi8(vtableF, vx2)); _mm256_storeu_si256((__m256i*) output, vy0); _mm256_storeu_si256((__m256i*) (output + 32), vy1); _mm256_storeu_si256((__m256i*) (output + 64), vy2); output += 96; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(_mm256_castsi256_si128(vtable0), vx); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable1), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable2), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable3), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable4), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable5), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable6), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable7), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable8), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable9), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableA), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableB), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableC), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableD), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableE), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableF), vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(_mm256_castsi256_si128(vtable0), vx); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable1), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable2), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable3), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable4), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable5), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable6), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable7), vx)); vx = _mm_sub_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable8), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtable9), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableA), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableB), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableC), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableD), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableE), vx)); vx = _mm_subs_epi8(vx, _mm256_castsi256_si128(voffset)); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(_mm256_castsi256_si128(vtableF), vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { _mm_storeu_si32(output, vy); vy = _mm_srli_epi64(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { _mm_storeu_si16(output, vy); vy = _mm_srli_epi32(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) _mm_extract_epi8(vy, 0); } } }
14,211
51.637037
98
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx512skx-vpshufb-x128.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx512skx-vpshufb.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx512skx_vpshufb_x128( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m512i vt0 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) table)); const __m512i vt1 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 16))); const __m512i vt2 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 32))); const __m512i vt3 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 48))); const __m512i vt4 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 64))); const __m512i vt5 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 80))); const __m512i vt6 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 96))); const __m512i vt7 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 112))); const __m512i vt8 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 128))); const __m512i vt9 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 144))); const __m512i vtA = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 160))); const __m512i vtB = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 176))); const __m512i vtC = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 192))); const __m512i vtD = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 208))); const __m512i vtE = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 224))); const __m512i vtF = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 240))); const __m512i vtable0 = vt0; const __m512i vtable1 = _mm512_xor_si512(vt0, vt1); const __m512i vtable2 = _mm512_xor_si512(vt1, vt2); const __m512i vtable3 = _mm512_xor_si512(vt2, vt3); const __m512i vtable4 = _mm512_xor_si512(vt3, vt4); const __m512i vtable5 = _mm512_xor_si512(vt4, vt5); const __m512i vtable6 = _mm512_xor_si512(vt5, vt6); const __m512i vtable7 = _mm512_xor_si512(vt6, vt7); const __m512i vtable8 = _mm512_xor_si512(_mm512_xor_si512(vt7, vt8), vtable0); const __m512i vtable9 = _mm512_xor_si512(_mm512_xor_si512(vt8, vt9), vtable1); const __m512i vtableA = _mm512_xor_si512(_mm512_xor_si512(vt9, vtA), vtable2); const __m512i vtableB = _mm512_xor_si512(_mm512_xor_si512(vtA, vtB), vtable3); const __m512i vtableC = _mm512_xor_si512(_mm512_xor_si512(vtB, vtC), vtable4); const __m512i vtableD = _mm512_xor_si512(_mm512_xor_si512(vtC, vtD), vtable5); const __m512i vtableE = _mm512_xor_si512(_mm512_xor_si512(vtD, vtE), vtable6); const __m512i vtableF = _mm512_xor_si512(_mm512_xor_si512(vtE, vtF), vtable7); const __m512i voffset = _mm512_set1_epi8(16); for (; batch >= 128 * sizeof(uint8_t); batch -= 128 * sizeof(uint8_t)) { __m512i vx0 = _mm512_loadu_si512(input); __m512i vx1 = _mm512_loadu_si512(input + 64); input += 128; __m512i vy0 = _mm512_shuffle_epi8(vtable0, vx0); __m512i vy1 = _mm512_shuffle_epi8(vtable0, vx1); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable1, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable1, vx1)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable2, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable2, vx1)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable3, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable3, vx1)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable4, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable4, vx1)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable5, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable5, vx1)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable6, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable6, vx1)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable7, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable7, vx1)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable8, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable8, vx1)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable9, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable9, vx1)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableA, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableA, vx1)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableB, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableB, vx1)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableC, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableC, vx1)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableD, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableD, vx1)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableE, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableE, vx1)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableF, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableF, vx1)); _mm512_storeu_si512(output, vy0); _mm512_storeu_si512(output + 64, vy1); output += 128; } for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m512i vx = _mm512_loadu_si512(input); input += 64; __m512i vy = _mm512_shuffle_epi8(vtable0, vx); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable1, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable2, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable3, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable4, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable5, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable6, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable7, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable8, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable9, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableA, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableB, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableC, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableD, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableE, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableF, vx)); _mm512_storeu_si512(output, vy); output += 64; } if XNN_UNLIKELY(batch != 0) { assert(batch < 64); const __mmask64 vmask = _cvtu64_mask64((uint64_t) ((UINT64_C(1) << batch) - UINT64_C(1))); __m512i vx = _mm512_maskz_loadu_epi8(vmask, input); __m512i vy = _mm512_shuffle_epi8(vtable0, vx); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable1, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable2, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable3, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable4, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable5, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable6, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable7, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable8, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable9, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableA, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableB, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableC, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableD, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableE, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableF, vx)); _mm512_mask_storeu_epi8(output, vmask, vy); } }
10,692
46.950673
94
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx512skx-vpshufb-x192.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx512skx-vpshufb.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx512skx_vpshufb_x192( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m512i vt0 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) table)); const __m512i vt1 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 16))); const __m512i vt2 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 32))); const __m512i vt3 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 48))); const __m512i vt4 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 64))); const __m512i vt5 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 80))); const __m512i vt6 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 96))); const __m512i vt7 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 112))); const __m512i vt8 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 128))); const __m512i vt9 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 144))); const __m512i vtA = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 160))); const __m512i vtB = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 176))); const __m512i vtC = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 192))); const __m512i vtD = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 208))); const __m512i vtE = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 224))); const __m512i vtF = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 240))); const __m512i vtable0 = vt0; const __m512i vtable1 = _mm512_xor_si512(vt0, vt1); const __m512i vtable2 = _mm512_xor_si512(vt1, vt2); const __m512i vtable3 = _mm512_xor_si512(vt2, vt3); const __m512i vtable4 = _mm512_xor_si512(vt3, vt4); const __m512i vtable5 = _mm512_xor_si512(vt4, vt5); const __m512i vtable6 = _mm512_xor_si512(vt5, vt6); const __m512i vtable7 = _mm512_xor_si512(vt6, vt7); const __m512i vtable8 = _mm512_xor_si512(_mm512_xor_si512(vt7, vt8), vtable0); const __m512i vtable9 = _mm512_xor_si512(_mm512_xor_si512(vt8, vt9), vtable1); const __m512i vtableA = _mm512_xor_si512(_mm512_xor_si512(vt9, vtA), vtable2); const __m512i vtableB = _mm512_xor_si512(_mm512_xor_si512(vtA, vtB), vtable3); const __m512i vtableC = _mm512_xor_si512(_mm512_xor_si512(vtB, vtC), vtable4); const __m512i vtableD = _mm512_xor_si512(_mm512_xor_si512(vtC, vtD), vtable5); const __m512i vtableE = _mm512_xor_si512(_mm512_xor_si512(vtD, vtE), vtable6); const __m512i vtableF = _mm512_xor_si512(_mm512_xor_si512(vtE, vtF), vtable7); const __m512i voffset = _mm512_set1_epi8(16); for (; batch >= 192 * sizeof(uint8_t); batch -= 192 * sizeof(uint8_t)) { __m512i vx0 = _mm512_loadu_si512(input); __m512i vx1 = _mm512_loadu_si512(input + 64); __m512i vx2 = _mm512_loadu_si512(input + 128); input += 192; __m512i vy0 = _mm512_shuffle_epi8(vtable0, vx0); __m512i vy1 = _mm512_shuffle_epi8(vtable0, vx1); __m512i vy2 = _mm512_shuffle_epi8(vtable0, vx2); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable1, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable1, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable1, vx2)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable2, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable2, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable2, vx2)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable3, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable3, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable3, vx2)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable4, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable4, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable4, vx2)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable5, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable5, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable5, vx2)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable6, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable6, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable6, vx2)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable7, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable7, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable7, vx2)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable8, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable8, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable8, vx2)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable9, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable9, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable9, vx2)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableA, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableA, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableA, vx2)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableB, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableB, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableB, vx2)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableC, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableC, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableC, vx2)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableD, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableD, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableD, vx2)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableE, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableE, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableE, vx2)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableF, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableF, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableF, vx2)); _mm512_storeu_si512(output, vy0); _mm512_storeu_si512(output + 64, vy1); _mm512_storeu_si512(output + 128, vy2); output += 192; } for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m512i vx = _mm512_loadu_si512(input); input += 64; __m512i vy = _mm512_shuffle_epi8(vtable0, vx); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable1, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable2, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable3, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable4, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable5, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable6, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable7, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable8, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable9, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableA, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableB, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableC, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableD, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableE, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableF, vx)); _mm512_storeu_si512(output, vy); output += 64; } if XNN_UNLIKELY(batch != 0) { assert(batch < 64); const __mmask64 vmask = _cvtu64_mask64((uint64_t) ((UINT64_C(1) << batch) - UINT64_C(1))); __m512i vx = _mm512_maskz_loadu_epi8(vmask, input); __m512i vy = _mm512_shuffle_epi8(vtable0, vx); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable1, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable2, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable3, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable4, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable5, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable6, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable7, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable8, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable9, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableA, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableB, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableC, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableD, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableE, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableF, vx)); _mm512_mask_storeu_epi8(output, vmask, vy); } }
12,482
47.761719
94
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx512skx-vpshufb-x256.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx512skx-vpshufb.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx512skx_vpshufb_x256( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m512i vt0 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) table)); const __m512i vt1 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 16))); const __m512i vt2 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 32))); const __m512i vt3 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 48))); const __m512i vt4 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 64))); const __m512i vt5 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 80))); const __m512i vt6 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 96))); const __m512i vt7 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 112))); const __m512i vt8 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 128))); const __m512i vt9 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 144))); const __m512i vtA = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 160))); const __m512i vtB = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 176))); const __m512i vtC = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 192))); const __m512i vtD = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 208))); const __m512i vtE = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 224))); const __m512i vtF = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 240))); const __m512i vtable0 = vt0; const __m512i vtable1 = _mm512_xor_si512(vt0, vt1); const __m512i vtable2 = _mm512_xor_si512(vt1, vt2); const __m512i vtable3 = _mm512_xor_si512(vt2, vt3); const __m512i vtable4 = _mm512_xor_si512(vt3, vt4); const __m512i vtable5 = _mm512_xor_si512(vt4, vt5); const __m512i vtable6 = _mm512_xor_si512(vt5, vt6); const __m512i vtable7 = _mm512_xor_si512(vt6, vt7); const __m512i vtable8 = _mm512_xor_si512(_mm512_xor_si512(vt7, vt8), vtable0); const __m512i vtable9 = _mm512_xor_si512(_mm512_xor_si512(vt8, vt9), vtable1); const __m512i vtableA = _mm512_xor_si512(_mm512_xor_si512(vt9, vtA), vtable2); const __m512i vtableB = _mm512_xor_si512(_mm512_xor_si512(vtA, vtB), vtable3); const __m512i vtableC = _mm512_xor_si512(_mm512_xor_si512(vtB, vtC), vtable4); const __m512i vtableD = _mm512_xor_si512(_mm512_xor_si512(vtC, vtD), vtable5); const __m512i vtableE = _mm512_xor_si512(_mm512_xor_si512(vtD, vtE), vtable6); const __m512i vtableF = _mm512_xor_si512(_mm512_xor_si512(vtE, vtF), vtable7); const __m512i voffset = _mm512_set1_epi8(16); for (; batch >= 256 * sizeof(uint8_t); batch -= 256 * sizeof(uint8_t)) { __m512i vx0 = _mm512_loadu_si512(input); __m512i vx1 = _mm512_loadu_si512(input + 64); __m512i vx2 = _mm512_loadu_si512(input + 128); __m512i vx3 = _mm512_loadu_si512(input + 192); input += 256; __m512i vy0 = _mm512_shuffle_epi8(vtable0, vx0); __m512i vy1 = _mm512_shuffle_epi8(vtable0, vx1); __m512i vy2 = _mm512_shuffle_epi8(vtable0, vx2); __m512i vy3 = _mm512_shuffle_epi8(vtable0, vx3); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vx3 = _mm512_sub_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable1, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable1, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable1, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtable1, vx3)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vx3 = _mm512_sub_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable2, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable2, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable2, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtable2, vx3)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vx3 = _mm512_sub_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable3, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable3, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable3, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtable3, vx3)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vx3 = _mm512_sub_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable4, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable4, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable4, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtable4, vx3)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vx3 = _mm512_sub_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable5, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable5, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable5, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtable5, vx3)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vx3 = _mm512_sub_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable6, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable6, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable6, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtable6, vx3)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vx3 = _mm512_sub_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable7, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable7, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable7, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtable7, vx3)); vx0 = _mm512_sub_epi8(vx0, voffset); vx1 = _mm512_sub_epi8(vx1, voffset); vx2 = _mm512_sub_epi8(vx2, voffset); vx3 = _mm512_sub_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable8, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable8, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable8, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtable8, vx3)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vx3 = _mm512_subs_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtable9, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtable9, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtable9, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtable9, vx3)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vx3 = _mm512_subs_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableA, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableA, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableA, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtableA, vx3)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vx3 = _mm512_subs_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableB, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableB, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableB, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtableB, vx3)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vx3 = _mm512_subs_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableC, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableC, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableC, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtableC, vx3)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vx3 = _mm512_subs_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableD, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableD, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableD, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtableD, vx3)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vx3 = _mm512_subs_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableE, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableE, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableE, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtableE, vx3)); vx0 = _mm512_subs_epi8(vx0, voffset); vx1 = _mm512_subs_epi8(vx1, voffset); vx2 = _mm512_subs_epi8(vx2, voffset); vx3 = _mm512_subs_epi8(vx3, voffset); vy0 = _mm512_xor_si512(vy0, _mm512_shuffle_epi8(vtableF, vx0)); vy1 = _mm512_xor_si512(vy1, _mm512_shuffle_epi8(vtableF, vx1)); vy2 = _mm512_xor_si512(vy2, _mm512_shuffle_epi8(vtableF, vx2)); vy3 = _mm512_xor_si512(vy3, _mm512_shuffle_epi8(vtableF, vx3)); _mm512_storeu_si512(output, vy0); _mm512_storeu_si512(output + 64, vy1); _mm512_storeu_si512(output + 128, vy2); _mm512_storeu_si512(output + 192, vy3); output += 256; } for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m512i vx = _mm512_loadu_si512(input); input += 64; __m512i vy = _mm512_shuffle_epi8(vtable0, vx); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable1, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable2, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable3, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable4, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable5, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable6, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable7, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable8, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable9, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableA, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableB, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableC, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableD, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableE, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableF, vx)); _mm512_storeu_si512(output, vy); output += 64; } if XNN_UNLIKELY(batch != 0) { assert(batch < 64); const __mmask64 vmask = _cvtu64_mask64((uint64_t) ((UINT64_C(1) << batch) - UINT64_C(1))); __m512i vx = _mm512_maskz_loadu_epi8(vmask, input); __m512i vy = _mm512_shuffle_epi8(vtable0, vx); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable1, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable2, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable3, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable4, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable5, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable6, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable7, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable8, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable9, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableA, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableB, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableC, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableD, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableE, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableF, vx)); _mm512_mask_storeu_epi8(output, vmask, vy); } }
14,272
48.387543
94
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx512skx-vpshufb-x64.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx512skx-vpshufb.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx512skx_vpshufb_x64( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m512i vt0 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) table)); const __m512i vt1 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 16))); const __m512i vt2 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 32))); const __m512i vt3 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 48))); const __m512i vt4 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 64))); const __m512i vt5 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 80))); const __m512i vt6 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 96))); const __m512i vt7 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 112))); const __m512i vt8 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 128))); const __m512i vt9 = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 144))); const __m512i vtA = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 160))); const __m512i vtB = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 176))); const __m512i vtC = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 192))); const __m512i vtD = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 208))); const __m512i vtE = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 224))); const __m512i vtF = _mm512_broadcast_i32x4(_mm_load_si128((const __m128i*) (table + 240))); const __m512i vtable0 = vt0; const __m512i vtable1 = _mm512_xor_si512(vt0, vt1); const __m512i vtable2 = _mm512_xor_si512(vt1, vt2); const __m512i vtable3 = _mm512_xor_si512(vt2, vt3); const __m512i vtable4 = _mm512_xor_si512(vt3, vt4); const __m512i vtable5 = _mm512_xor_si512(vt4, vt5); const __m512i vtable6 = _mm512_xor_si512(vt5, vt6); const __m512i vtable7 = _mm512_xor_si512(vt6, vt7); const __m512i vtable8 = _mm512_xor_si512(_mm512_xor_si512(vt7, vt8), vtable0); const __m512i vtable9 = _mm512_xor_si512(_mm512_xor_si512(vt8, vt9), vtable1); const __m512i vtableA = _mm512_xor_si512(_mm512_xor_si512(vt9, vtA), vtable2); const __m512i vtableB = _mm512_xor_si512(_mm512_xor_si512(vtA, vtB), vtable3); const __m512i vtableC = _mm512_xor_si512(_mm512_xor_si512(vtB, vtC), vtable4); const __m512i vtableD = _mm512_xor_si512(_mm512_xor_si512(vtC, vtD), vtable5); const __m512i vtableE = _mm512_xor_si512(_mm512_xor_si512(vtD, vtE), vtable6); const __m512i vtableF = _mm512_xor_si512(_mm512_xor_si512(vtE, vtF), vtable7); const __m512i voffset = _mm512_set1_epi8(16); for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m512i vx = _mm512_loadu_si512(input); input += 64; __m512i vy = _mm512_shuffle_epi8(vtable0, vx); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable1, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable2, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable3, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable4, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable5, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable6, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable7, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable8, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable9, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableA, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableB, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableC, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableD, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableE, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableF, vx)); _mm512_storeu_si512(output, vy); output += 64; } if XNN_UNLIKELY(batch != 0) { assert(batch < 64); const __mmask64 vmask = _cvtu64_mask64((uint64_t) ((UINT64_C(1) << batch) - UINT64_C(1))); __m512i vx = _mm512_maskz_loadu_epi8(vmask, input); __m512i vy = _mm512_shuffle_epi8(vtable0, vx); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable1, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable2, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable3, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable4, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable5, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable6, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable7, vx)); vx = _mm512_sub_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable8, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtable9, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableA, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableB, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableC, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableD, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableE, vx)); vx = _mm512_subs_epi8(vx, voffset); vy = _mm512_xor_si512(vy, _mm512_shuffle_epi8(vtableF, vx)); _mm512_mask_storeu_epi8(output, vmask, vy); } }
7,005
46.020134
94
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx512vbmi-vpermx2b-x128.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx512vbmi-vpermx2b.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx512vbmi_vpermx2b_x128( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m512i vtable0 = _mm512_loadu_si512(table); const __m512i vtable1 = _mm512_loadu_si512(table + 64); const __m512i vtable2 = _mm512_loadu_si512(table + 128); const __m512i vtable3 = _mm512_loadu_si512(table + 192); for (; batch >= 128 * sizeof(uint8_t); batch -= 128 * sizeof(uint8_t)) { const __m512i vx0 = _mm512_loadu_si512(input); const __m512i vx1 = _mm512_loadu_si512(input + 64); input += 128; __m512i vy0 = _mm512_permutex2var_epi8(vtable0, vx0, vtable1); const __mmask64 vm0 = _mm512_movepi8_mask(vx0); __m512i vy1 = _mm512_permutex2var_epi8(vtable0, vx1, vtable1); const __mmask64 vm1 = _mm512_movepi8_mask(vx1); const __m512i vt0 = _mm512_permutex2var_epi8(vtable2, vx0, vtable3); const __m512i vt1 = _mm512_permutex2var_epi8(vtable2, vx1, vtable3); vy0 = _mm512_mask_mov_epi8(vy0, vm0, vt0); vy1 = _mm512_mask_mov_epi8(vy1, vm1, vt1); _mm512_storeu_si512(output, vy0); _mm512_storeu_si512(output + 64, vy1); output += 128; } for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m512i vx = _mm512_loadu_si512(input); input += 64; __m512i vy = _mm512_permutex2var_epi8(vtable0, vx, vtable1); const __mmask64 vm = _mm512_movepi8_mask(vx); const __m512i vt = _mm512_permutex2var_epi8(vtable2, vx, vtable3); vy = _mm512_mask_mov_epi8(vy, vm, vt); _mm512_storeu_si512(output, vy); output += 64; } if XNN_UNLIKELY(batch != 0) { assert(batch < 64); const __mmask64 vmask = _cvtu64_mask64((uint64_t) ((UINT64_C(1) << batch) - UINT64_C(1))); __m512i vx = _mm512_maskz_loadu_epi8(vmask, input); __m512i vy = _mm512_maskz_permutex2var_epi8(vmask, vtable0, vx, vtable1); const __mmask64 vm = _mm512_movepi8_mask(vx); const __m512i vt = _mm512_maskz_permutex2var_epi8(vmask, vtable2, vx, vtable3); vy = _mm512_mask_mov_epi8(vy, vm, vt); _mm512_mask_storeu_epi8(output, vmask, vy); } }
2,673
32.012346
94
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx512vbmi-vpermx2b-x192.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx512vbmi-vpermx2b.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx512vbmi_vpermx2b_x192( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m512i vtable0 = _mm512_loadu_si512(table); const __m512i vtable1 = _mm512_loadu_si512(table + 64); const __m512i vtable2 = _mm512_loadu_si512(table + 128); const __m512i vtable3 = _mm512_loadu_si512(table + 192); for (; batch >= 192 * sizeof(uint8_t); batch -= 192 * sizeof(uint8_t)) { const __m512i vx0 = _mm512_loadu_si512(input); const __m512i vx1 = _mm512_loadu_si512(input + 64); const __m512i vx2 = _mm512_loadu_si512(input + 128); input += 192; __m512i vy0 = _mm512_permutex2var_epi8(vtable0, vx0, vtable1); const __mmask64 vm0 = _mm512_movepi8_mask(vx0); __m512i vy1 = _mm512_permutex2var_epi8(vtable0, vx1, vtable1); const __mmask64 vm1 = _mm512_movepi8_mask(vx1); __m512i vy2 = _mm512_permutex2var_epi8(vtable0, vx2, vtable1); const __mmask64 vm2 = _mm512_movepi8_mask(vx2); const __m512i vt0 = _mm512_permutex2var_epi8(vtable2, vx0, vtable3); const __m512i vt1 = _mm512_permutex2var_epi8(vtable2, vx1, vtable3); const __m512i vt2 = _mm512_permutex2var_epi8(vtable2, vx2, vtable3); vy0 = _mm512_mask_mov_epi8(vy0, vm0, vt0); vy1 = _mm512_mask_mov_epi8(vy1, vm1, vt1); vy2 = _mm512_mask_mov_epi8(vy2, vm2, vt2); _mm512_storeu_si512(output, vy0); _mm512_storeu_si512(output + 64, vy1); _mm512_storeu_si512(output + 128, vy2); output += 192; } for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m512i vx = _mm512_loadu_si512(input); input += 64; __m512i vy = _mm512_permutex2var_epi8(vtable0, vx, vtable1); const __mmask64 vm = _mm512_movepi8_mask(vx); const __m512i vt = _mm512_permutex2var_epi8(vtable2, vx, vtable3); vy = _mm512_mask_mov_epi8(vy, vm, vt); _mm512_storeu_si512(output, vy); output += 64; } if XNN_UNLIKELY(batch != 0) { assert(batch < 64); const __mmask64 vmask = _cvtu64_mask64((uint64_t) ((UINT64_C(1) << batch) - UINT64_C(1))); __m512i vx = _mm512_maskz_loadu_epi8(vmask, input); __m512i vy = _mm512_maskz_permutex2var_epi8(vmask, vtable0, vx, vtable1); const __mmask64 vm = _mm512_movepi8_mask(vx); const __m512i vt = _mm512_maskz_permutex2var_epi8(vmask, vtable2, vx, vtable3); vy = _mm512_mask_mov_epi8(vy, vm, vt); _mm512_mask_storeu_epi8(output, vmask, vy); } }
3,013
33.643678
94
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx512vbmi-vpermx2b-x256.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx512vbmi-vpermx2b.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx512vbmi_vpermx2b_x256( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m512i vtable0 = _mm512_loadu_si512(table); const __m512i vtable1 = _mm512_loadu_si512(table + 64); const __m512i vtable2 = _mm512_loadu_si512(table + 128); const __m512i vtable3 = _mm512_loadu_si512(table + 192); for (; batch >= 256 * sizeof(uint8_t); batch -= 256 * sizeof(uint8_t)) { const __m512i vx0 = _mm512_loadu_si512(input); const __m512i vx1 = _mm512_loadu_si512(input + 64); const __m512i vx2 = _mm512_loadu_si512(input + 128); const __m512i vx3 = _mm512_loadu_si512(input + 192); input += 256; __m512i vy0 = _mm512_permutex2var_epi8(vtable0, vx0, vtable1); const __mmask64 vm0 = _mm512_movepi8_mask(vx0); __m512i vy1 = _mm512_permutex2var_epi8(vtable0, vx1, vtable1); const __mmask64 vm1 = _mm512_movepi8_mask(vx1); __m512i vy2 = _mm512_permutex2var_epi8(vtable0, vx2, vtable1); const __mmask64 vm2 = _mm512_movepi8_mask(vx2); __m512i vy3 = _mm512_permutex2var_epi8(vtable0, vx3, vtable1); const __mmask64 vm3 = _mm512_movepi8_mask(vx3); const __m512i vt0 = _mm512_permutex2var_epi8(vtable2, vx0, vtable3); const __m512i vt1 = _mm512_permutex2var_epi8(vtable2, vx1, vtable3); const __m512i vt2 = _mm512_permutex2var_epi8(vtable2, vx2, vtable3); const __m512i vt3 = _mm512_permutex2var_epi8(vtable2, vx3, vtable3); vy0 = _mm512_mask_mov_epi8(vy0, vm0, vt0); vy1 = _mm512_mask_mov_epi8(vy1, vm1, vt1); vy2 = _mm512_mask_mov_epi8(vy2, vm2, vt2); vy3 = _mm512_mask_mov_epi8(vy3, vm3, vt3); _mm512_storeu_si512(output, vy0); _mm512_storeu_si512(output + 64, vy1); _mm512_storeu_si512(output + 128, vy2); _mm512_storeu_si512(output + 192, vy3); output += 256; } for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m512i vx = _mm512_loadu_si512(input); input += 64; __m512i vy = _mm512_permutex2var_epi8(vtable0, vx, vtable1); const __mmask64 vm = _mm512_movepi8_mask(vx); const __m512i vt = _mm512_permutex2var_epi8(vtable2, vx, vtable3); vy = _mm512_mask_mov_epi8(vy, vm, vt); _mm512_storeu_si512(output, vy); output += 64; } if XNN_UNLIKELY(batch != 0) { assert(batch < 64); const __mmask64 vmask = _cvtu64_mask64((uint64_t) ((UINT64_C(1) << batch) - UINT64_C(1))); __m512i vx = _mm512_maskz_loadu_epi8(vmask, input); __m512i vy = _mm512_maskz_permutex2var_epi8(vmask, vtable0, vx, vtable1); const __mmask64 vm = _mm512_movepi8_mask(vx); const __m512i vt = _mm512_maskz_permutex2var_epi8(vmask, vtable2, vx, vtable3); vy = _mm512_mask_mov_epi8(vy, vm, vt); _mm512_mask_storeu_epi8(output, vmask, vy); } }
3,353
35.064516
94
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-avx512vbmi-vpermx2b-x64.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/avx512vbmi-vpermx2b.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <immintrin.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__avx512vbmi_vpermx2b_x64( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m512i vtable0 = _mm512_loadu_si512(table); const __m512i vtable1 = _mm512_loadu_si512(table + 64); const __m512i vtable2 = _mm512_loadu_si512(table + 128); const __m512i vtable3 = _mm512_loadu_si512(table + 192); for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { __m512i vx = _mm512_loadu_si512(input); input += 64; __m512i vy = _mm512_permutex2var_epi8(vtable0, vx, vtable1); const __mmask64 vm = _mm512_movepi8_mask(vx); const __m512i vt = _mm512_permutex2var_epi8(vtable2, vx, vtable3); vy = _mm512_mask_mov_epi8(vy, vm, vt); _mm512_storeu_si512(output, vy); output += 64; } if XNN_UNLIKELY(batch != 0) { assert(batch < 64); const __mmask64 vmask = _cvtu64_mask64((uint64_t) ((UINT64_C(1) << batch) - UINT64_C(1))); __m512i vx = _mm512_maskz_loadu_epi8(vmask, input); __m512i vy = _mm512_maskz_permutex2var_epi8(vmask, vtable0, vx, vtable1); const __mmask64 vm = _mm512_movepi8_mask(vx); const __m512i vt = _mm512_maskz_permutex2var_epi8(vmask, vtable2, vx, vtable3); vy = _mm512_mask_mov_epi8(vy, vm, vt); _mm512_mask_storeu_epi8(output, vmask, vy); } }
1,886
29.934426
94
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-scalar-x1.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__scalar_x1( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); do { const size_t vx = (size_t) *input++; const uint32_t vt = (uint32_t) table[vx]; *output++ = (uint8_t) vt; batch -= sizeof(uint8_t); } while (batch != 0); }
801
22.588235
72
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-scalar-x16.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__scalar_x16( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { const size_t vx0 = (size_t) input[0]; const size_t vx1 = (size_t) input[1]; const size_t vx2 = (size_t) input[2]; const size_t vx3 = (size_t) input[3]; const size_t vx4 = (size_t) input[4]; const size_t vx5 = (size_t) input[5]; const size_t vx6 = (size_t) input[6]; const size_t vx7 = (size_t) input[7]; const size_t vx8 = (size_t) input[8]; const size_t vx9 = (size_t) input[9]; const size_t vx10 = (size_t) input[10]; const size_t vx11 = (size_t) input[11]; const size_t vx12 = (size_t) input[12]; const size_t vx13 = (size_t) input[13]; const size_t vx14 = (size_t) input[14]; const size_t vx15 = (size_t) input[15]; input += 16; const uint32_t vt0 = (uint32_t) table[vx0]; const uint32_t vt1 = (uint32_t) table[vx1]; const uint32_t vt2 = (uint32_t) table[vx2]; const uint32_t vt3 = (uint32_t) table[vx3]; const uint32_t vt4 = (uint32_t) table[vx4]; const uint32_t vt5 = (uint32_t) table[vx5]; const uint32_t vt6 = (uint32_t) table[vx6]; const uint32_t vt7 = (uint32_t) table[vx7]; const uint32_t vt8 = (uint32_t) table[vx8]; const uint32_t vt9 = (uint32_t) table[vx9]; const uint32_t vt10 = (uint32_t) table[vx10]; const uint32_t vt11 = (uint32_t) table[vx11]; const uint32_t vt12 = (uint32_t) table[vx12]; const uint32_t vt13 = (uint32_t) table[vx13]; const uint32_t vt14 = (uint32_t) table[vx14]; const uint32_t vt15 = (uint32_t) table[vx15]; output[0] = (uint8_t) vt0; output[1] = (uint8_t) vt1; output[2] = (uint8_t) vt2; output[3] = (uint8_t) vt3; output[4] = (uint8_t) vt4; output[5] = (uint8_t) vt5; output[6] = (uint8_t) vt6; output[7] = (uint8_t) vt7; output[8] = (uint8_t) vt8; output[9] = (uint8_t) vt9; output[10] = (uint8_t) vt10; output[11] = (uint8_t) vt11; output[12] = (uint8_t) vt12; output[13] = (uint8_t) vt13; output[14] = (uint8_t) vt14; output[15] = (uint8_t) vt15; output += 16; } if XNN_UNLIKELY(batch != 0) { do { const size_t vx = (size_t) *input++; const uint32_t vt = (uint32_t) table[vx]; *output++ = (uint8_t) vt; batch -= sizeof(uint8_t); } while (batch != 0); } }
2,936
31.633333
72
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-scalar-x2.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__scalar_x2( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); for (; batch >= 2 * sizeof(uint8_t); batch -= 2 * sizeof(uint8_t)) { const size_t vx0 = (size_t) input[0]; const size_t vx1 = (size_t) input[1]; input += 2; const uint32_t vt0 = (uint32_t) table[vx0]; const uint32_t vt1 = (uint32_t) table[vx1]; output[0] = (uint8_t) vt0; output[1] = (uint8_t) vt1; output += 2; } if XNN_UNLIKELY(batch != 0) { const size_t vx = (size_t) *input; const uint32_t vt = (uint32_t) table[vx]; *output = (uint8_t) vt; } }
1,124
24
72
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-scalar-x4.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__scalar_x4( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); for (; batch >= 4 * sizeof(uint8_t); batch -= 4 * sizeof(uint8_t)) { const size_t vx0 = (size_t) input[0]; const size_t vx1 = (size_t) input[1]; const size_t vx2 = (size_t) input[2]; const size_t vx3 = (size_t) input[3]; input += 4; const uint32_t vt0 = (uint32_t) table[vx0]; const uint32_t vt1 = (uint32_t) table[vx1]; const uint32_t vt2 = (uint32_t) table[vx2]; const uint32_t vt3 = (uint32_t) table[vx3]; output[0] = (uint8_t) vt0; output[1] = (uint8_t) vt1; output[2] = (uint8_t) vt2; output[3] = (uint8_t) vt3; output += 4; } if XNN_UNLIKELY(batch != 0) { do { const size_t vx = (size_t) *input++; const uint32_t vt = (uint32_t) table[vx]; *output++ = (uint8_t) vt; batch -= sizeof(uint8_t); } while (batch != 0); } }
1,443
25.740741
72
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-scalar-x8.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__scalar_x8( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); for (; batch >= 8 * sizeof(uint8_t); batch -= 8 * sizeof(uint8_t)) { const size_t vx0 = (size_t) input[0]; const size_t vx1 = (size_t) input[1]; const size_t vx2 = (size_t) input[2]; const size_t vx3 = (size_t) input[3]; const size_t vx4 = (size_t) input[4]; const size_t vx5 = (size_t) input[5]; const size_t vx6 = (size_t) input[6]; const size_t vx7 = (size_t) input[7]; input += 8; const uint32_t vt0 = (uint32_t) table[vx0]; const uint32_t vt1 = (uint32_t) table[vx1]; const uint32_t vt2 = (uint32_t) table[vx2]; const uint32_t vt3 = (uint32_t) table[vx3]; const uint32_t vt4 = (uint32_t) table[vx4]; const uint32_t vt5 = (uint32_t) table[vx5]; const uint32_t vt6 = (uint32_t) table[vx6]; const uint32_t vt7 = (uint32_t) table[vx7]; output[0] = (uint8_t) vt0; output[1] = (uint8_t) vt1; output[2] = (uint8_t) vt2; output[3] = (uint8_t) vt3; output[4] = (uint8_t) vt4; output[5] = (uint8_t) vt5; output[6] = (uint8_t) vt6; output[7] = (uint8_t) vt7; output += 8; } if XNN_UNLIKELY(batch != 0) { do { const size_t vx = (size_t) *input++; const uint32_t vt = (uint32_t) table[vx]; *output++ = (uint8_t) vt; batch -= sizeof(uint8_t); } while (batch != 0); } }
1,927
28.212121
72
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-ssse3-x16.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/ssse3.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <tmmintrin.h> #include <xnnpack/common.h> #include <xnnpack/lut.h> #include <xnnpack/unaligned.h> void xnn_x8_lut_ukernel__ssse3_x16( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m128i vt0 = _mm_load_si128((const __m128i*) table); const __m128i vt1 = _mm_load_si128((const __m128i*) (table + 16)); const __m128i vt2 = _mm_load_si128((const __m128i*) (table + 32)); const __m128i vt3 = _mm_load_si128((const __m128i*) (table + 48)); const __m128i vt4 = _mm_load_si128((const __m128i*) (table + 64)); const __m128i vt5 = _mm_load_si128((const __m128i*) (table + 80)); const __m128i vt6 = _mm_load_si128((const __m128i*) (table + 96)); const __m128i vt7 = _mm_load_si128((const __m128i*) (table + 112)); const __m128i vt8 = _mm_load_si128((const __m128i*) (table + 128)); const __m128i vt9 = _mm_load_si128((const __m128i*) (table + 144)); const __m128i vtA = _mm_load_si128((const __m128i*) (table + 160)); const __m128i vtB = _mm_load_si128((const __m128i*) (table + 176)); const __m128i vtC = _mm_load_si128((const __m128i*) (table + 192)); const __m128i vtD = _mm_load_si128((const __m128i*) (table + 208)); const __m128i vtE = _mm_load_si128((const __m128i*) (table + 224)); const __m128i vtF = _mm_load_si128((const __m128i*) (table + 240)); const __m128i vtable0 = vt0; const __m128i vtable1 = _mm_xor_si128(vt0, vt1); const __m128i vtable2 = _mm_xor_si128(vt1, vt2); const __m128i vtable3 = _mm_xor_si128(vt2, vt3); const __m128i vtable4 = _mm_xor_si128(vt3, vt4); const __m128i vtable5 = _mm_xor_si128(vt4, vt5); const __m128i vtable6 = _mm_xor_si128(vt5, vt6); const __m128i vtable7 = _mm_xor_si128(vt6, vt7); const __m128i vtable8 = _mm_xor_si128(_mm_xor_si128(vt7, vt8), vtable0); const __m128i vtable9 = _mm_xor_si128(_mm_xor_si128(vt8, vt9), vtable1); const __m128i vtableA = _mm_xor_si128(_mm_xor_si128(vt9, vtA), vtable2); const __m128i vtableB = _mm_xor_si128(_mm_xor_si128(vtA, vtB), vtable3); const __m128i vtableC = _mm_xor_si128(_mm_xor_si128(vtB, vtC), vtable4); const __m128i vtableD = _mm_xor_si128(_mm_xor_si128(vtC, vtD), vtable5); const __m128i vtableE = _mm_xor_si128(_mm_xor_si128(vtD, vtE), vtable6); const __m128i vtableF = _mm_xor_si128(_mm_xor_si128(vtE, vtF), vtable7); const __m128i voffset = _mm_set1_epi8(16); for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vy)); vy = _mm_srli_epi64(vy, 32); output += 4; } uint32_t vy_lo = (uint32_t) _mm_cvtsi128_si32(vy); if (batch & (2 * sizeof(uint8_t))) { unaligned_store_u16(output, (uint16_t) vy_lo); vy_lo >>= 16; output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) vy_lo; } } }
6,691
39.804878
74
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-ssse3-x32.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/ssse3.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <tmmintrin.h> #include <xnnpack/common.h> #include <xnnpack/lut.h> #include <xnnpack/unaligned.h> void xnn_x8_lut_ukernel__ssse3_x32( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const __m128i vt0 = _mm_load_si128((const __m128i*) table); const __m128i vt1 = _mm_load_si128((const __m128i*) (table + 16)); const __m128i vt2 = _mm_load_si128((const __m128i*) (table + 32)); const __m128i vt3 = _mm_load_si128((const __m128i*) (table + 48)); const __m128i vt4 = _mm_load_si128((const __m128i*) (table + 64)); const __m128i vt5 = _mm_load_si128((const __m128i*) (table + 80)); const __m128i vt6 = _mm_load_si128((const __m128i*) (table + 96)); const __m128i vt7 = _mm_load_si128((const __m128i*) (table + 112)); const __m128i vt8 = _mm_load_si128((const __m128i*) (table + 128)); const __m128i vt9 = _mm_load_si128((const __m128i*) (table + 144)); const __m128i vtA = _mm_load_si128((const __m128i*) (table + 160)); const __m128i vtB = _mm_load_si128((const __m128i*) (table + 176)); const __m128i vtC = _mm_load_si128((const __m128i*) (table + 192)); const __m128i vtD = _mm_load_si128((const __m128i*) (table + 208)); const __m128i vtE = _mm_load_si128((const __m128i*) (table + 224)); const __m128i vtF = _mm_load_si128((const __m128i*) (table + 240)); const __m128i vtable0 = vt0; const __m128i vtable1 = _mm_xor_si128(vt0, vt1); const __m128i vtable2 = _mm_xor_si128(vt1, vt2); const __m128i vtable3 = _mm_xor_si128(vt2, vt3); const __m128i vtable4 = _mm_xor_si128(vt3, vt4); const __m128i vtable5 = _mm_xor_si128(vt4, vt5); const __m128i vtable6 = _mm_xor_si128(vt5, vt6); const __m128i vtable7 = _mm_xor_si128(vt6, vt7); const __m128i vtable8 = _mm_xor_si128(_mm_xor_si128(vt7, vt8), vtable0); const __m128i vtable9 = _mm_xor_si128(_mm_xor_si128(vt8, vt9), vtable1); const __m128i vtableA = _mm_xor_si128(_mm_xor_si128(vt9, vtA), vtable2); const __m128i vtableB = _mm_xor_si128(_mm_xor_si128(vtA, vtB), vtable3); const __m128i vtableC = _mm_xor_si128(_mm_xor_si128(vtB, vtC), vtable4); const __m128i vtableD = _mm_xor_si128(_mm_xor_si128(vtC, vtD), vtable5); const __m128i vtableE = _mm_xor_si128(_mm_xor_si128(vtD, vtE), vtable6); const __m128i vtableF = _mm_xor_si128(_mm_xor_si128(vtE, vtF), vtable7); const __m128i voffset = _mm_set1_epi8(16); for (; batch >= 32 * sizeof(uint8_t); batch -= 32 * sizeof(uint8_t)) { __m128i vx0 = _mm_loadu_si128((const __m128i*) input); __m128i vx1 = _mm_loadu_si128((const __m128i*) (input + 16)); input += 32; __m128i vy0 = _mm_shuffle_epi8(vtable0, vx0); __m128i vy1 = _mm_shuffle_epi8(vtable0, vx1); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable1, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable1, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable2, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable2, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable3, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable3, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable4, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable4, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable5, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable5, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable6, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable6, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable7, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable7, vx1)); vx0 = _mm_sub_epi8(vx0, voffset); vx1 = _mm_sub_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable8, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable8, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtable9, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtable9, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableA, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableA, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableB, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableB, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableC, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableC, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableD, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableD, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableE, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableE, vx1)); vx0 = _mm_subs_epi8(vx0, voffset); vx1 = _mm_subs_epi8(vx1, voffset); vy0 = _mm_xor_si128(vy0, _mm_shuffle_epi8(vtableF, vx0)); vy1 = _mm_xor_si128(vy1, _mm_shuffle_epi8(vtableF, vx1)); _mm_storeu_si128((__m128i*) output, vy0); _mm_storeu_si128((__m128i*) (output + 16), vy1); output += 32; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { __m128i vx = _mm_loadu_si128((const __m128i*) input); input += 16; __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); _mm_storeu_si128((__m128i*) output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { __m128i vx = _mm_loadu_si128((const __m128i*) input); __m128i vy = _mm_shuffle_epi8(vtable0, vx); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable1, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable2, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable3, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable4, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable5, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable6, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable7, vx)); vx = _mm_sub_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable8, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtable9, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableA, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableB, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableC, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableD, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableE, vx)); vx = _mm_subs_epi8(vx, voffset); vy = _mm_xor_si128(vy, _mm_shuffle_epi8(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { _mm_storel_epi64((__m128i*) output, vy); vy = _mm_unpackhi_epi64(vy, vy); output += 8; } if (batch & (4 * sizeof(uint8_t))) { unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vy)); vy = _mm_srli_epi64(vy, 32); output += 4; } uint32_t vy_lo = (uint32_t) _mm_cvtsi128_si32(vy); if (batch & (2 * sizeof(uint8_t))) { unaligned_store_u16(output, (uint16_t) vy_lo); vy_lo >>= 16; output += 2; } if (batch & (1 * sizeof(uint8_t))) { *output = (uint8_t) vy_lo; } } }
10,145
41.630252
74
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-wasmpshufb-x16.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/wasmpshufb.c.in // Generator: tools/xngen // // Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <wasm_simd128.h> #include <xnnpack/common.h> #include <xnnpack/lut.h> void xnn_x8_lut_ukernel__wasmpshufb_x16( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const v128_t vt0 = wasm_v128_load(table); const v128_t vt1 = wasm_v128_load(table + 16); const v128_t vt2 = wasm_v128_load(table + 32); const v128_t vt3 = wasm_v128_load(table + 48); const v128_t vt4 = wasm_v128_load(table + 64); const v128_t vt5 = wasm_v128_load(table + 80); const v128_t vt6 = wasm_v128_load(table + 96); const v128_t vt7 = wasm_v128_load(table + 112); const v128_t vt8 = wasm_v128_load(table + 128); const v128_t vt9 = wasm_v128_load(table + 144); const v128_t vtA = wasm_v128_load(table + 160); const v128_t vtB = wasm_v128_load(table + 176); const v128_t vtC = wasm_v128_load(table + 192); const v128_t vtD = wasm_v128_load(table + 208); const v128_t vtE = wasm_v128_load(table + 224); const v128_t vtF = wasm_v128_load(table + 240); const v128_t vtable0 = vt0; const v128_t vtable1 = wasm_v128_xor(vt0, vt1); const v128_t vtable2 = wasm_v128_xor(vt1, vt2); const v128_t vtable3 = wasm_v128_xor(vt2, vt3); const v128_t vtable4 = wasm_v128_xor(vt3, vt4); const v128_t vtable5 = wasm_v128_xor(vt4, vt5); const v128_t vtable6 = wasm_v128_xor(vt5, vt6); const v128_t vtable7 = wasm_v128_xor(vt6, vt7); const v128_t vtable8 = wasm_v128_xor(wasm_v128_xor(vt7, vt8), vtable0); const v128_t vtable9 = wasm_v128_xor(wasm_v128_xor(vt8, vt9), vtable1); const v128_t vtableA = wasm_v128_xor(wasm_v128_xor(vt9, vtA), vtable2); const v128_t vtableB = wasm_v128_xor(wasm_v128_xor(vtA, vtB), vtable3); const v128_t vtableC = wasm_v128_xor(wasm_v128_xor(vtB, vtC), vtable4); const v128_t vtableD = wasm_v128_xor(wasm_v128_xor(vtC, vtD), vtable5); const v128_t vtableE = wasm_v128_xor(wasm_v128_xor(vtD, vtE), vtable6); const v128_t vtableF = wasm_v128_xor(wasm_v128_xor(vtE, vtF), vtable7); const v128_t voffset = wasm_i8x16_const_splat(16); for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { v128_t vx = wasm_v128_load(input); input += 16; v128_t vy = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx)); wasm_v128_store(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { v128_t vx = wasm_v128_load((const v128_t*) input); v128_t vy = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { wasm_v128_store64_lane(output, vy, 0); vy = wasm_v64x2_shuffle(vy, vy, 1, 1); output += 8; } if (batch & (4 * sizeof(uint8_t))) { wasm_v128_store32_lane(output, vy, 0); vy = wasm_u64x2_shr(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { wasm_v128_store16_lane(output, vy, 0); vy = wasm_u32x4_shr(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { wasm_v128_store8_lane(output, vy, 0); } } }
6,996
42.191358
78
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-wasmpshufb-x32.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/wasmpshufb.c.in // Generator: tools/xngen // // Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <wasm_simd128.h> #include <xnnpack/common.h> #include <xnnpack/lut.h> void xnn_x8_lut_ukernel__wasmpshufb_x32( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const v128_t vt0 = wasm_v128_load(table); const v128_t vt1 = wasm_v128_load(table + 16); const v128_t vt2 = wasm_v128_load(table + 32); const v128_t vt3 = wasm_v128_load(table + 48); const v128_t vt4 = wasm_v128_load(table + 64); const v128_t vt5 = wasm_v128_load(table + 80); const v128_t vt6 = wasm_v128_load(table + 96); const v128_t vt7 = wasm_v128_load(table + 112); const v128_t vt8 = wasm_v128_load(table + 128); const v128_t vt9 = wasm_v128_load(table + 144); const v128_t vtA = wasm_v128_load(table + 160); const v128_t vtB = wasm_v128_load(table + 176); const v128_t vtC = wasm_v128_load(table + 192); const v128_t vtD = wasm_v128_load(table + 208); const v128_t vtE = wasm_v128_load(table + 224); const v128_t vtF = wasm_v128_load(table + 240); const v128_t vtable0 = vt0; const v128_t vtable1 = wasm_v128_xor(vt0, vt1); const v128_t vtable2 = wasm_v128_xor(vt1, vt2); const v128_t vtable3 = wasm_v128_xor(vt2, vt3); const v128_t vtable4 = wasm_v128_xor(vt3, vt4); const v128_t vtable5 = wasm_v128_xor(vt4, vt5); const v128_t vtable6 = wasm_v128_xor(vt5, vt6); const v128_t vtable7 = wasm_v128_xor(vt6, vt7); const v128_t vtable8 = wasm_v128_xor(wasm_v128_xor(vt7, vt8), vtable0); const v128_t vtable9 = wasm_v128_xor(wasm_v128_xor(vt8, vt9), vtable1); const v128_t vtableA = wasm_v128_xor(wasm_v128_xor(vt9, vtA), vtable2); const v128_t vtableB = wasm_v128_xor(wasm_v128_xor(vtA, vtB), vtable3); const v128_t vtableC = wasm_v128_xor(wasm_v128_xor(vtB, vtC), vtable4); const v128_t vtableD = wasm_v128_xor(wasm_v128_xor(vtC, vtD), vtable5); const v128_t vtableE = wasm_v128_xor(wasm_v128_xor(vtD, vtE), vtable6); const v128_t vtableF = wasm_v128_xor(wasm_v128_xor(vtE, vtF), vtable7); const v128_t voffset = wasm_i8x16_const_splat(16); for (; batch >= 32 * sizeof(uint8_t); batch -= 32 * sizeof(uint8_t)) { v128_t vx0 = wasm_v128_load((const v128_t*) input); v128_t vx1 = wasm_v128_load((const v128_t*) (input + 16)); input += 32; v128_t vy0 = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx0); v128_t vy1 = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx1); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx1)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx1)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx1)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx1)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx1)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx1)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx1)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx1)); wasm_v128_store(output, vy0); wasm_v128_store(output + 16, vy1); output += 32; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { v128_t vx = wasm_v128_load(input); input += 16; v128_t vy = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx)); wasm_v128_store(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { v128_t vx = wasm_v128_load((const v128_t*) input); v128_t vy = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { wasm_v128_store64_lane(output, vy, 0); vy = wasm_v64x2_shuffle(vy, vy, 1, 1); output += 8; } if (batch & (4 * sizeof(uint8_t))) { wasm_v128_store32_lane(output, vy, 0); vy = wasm_u64x2_shr(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { wasm_v128_store16_lane(output, vy, 0); vy = wasm_u32x4_shr(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { wasm_v128_store8_lane(output, vy, 0); } } }
11,158
46.283898
81
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-wasmpshufb-x48.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/wasmpshufb.c.in // Generator: tools/xngen // // Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <wasm_simd128.h> #include <xnnpack/common.h> #include <xnnpack/lut.h> void xnn_x8_lut_ukernel__wasmpshufb_x48( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const v128_t vt0 = wasm_v128_load(table); const v128_t vt1 = wasm_v128_load(table + 16); const v128_t vt2 = wasm_v128_load(table + 32); const v128_t vt3 = wasm_v128_load(table + 48); const v128_t vt4 = wasm_v128_load(table + 64); const v128_t vt5 = wasm_v128_load(table + 80); const v128_t vt6 = wasm_v128_load(table + 96); const v128_t vt7 = wasm_v128_load(table + 112); const v128_t vt8 = wasm_v128_load(table + 128); const v128_t vt9 = wasm_v128_load(table + 144); const v128_t vtA = wasm_v128_load(table + 160); const v128_t vtB = wasm_v128_load(table + 176); const v128_t vtC = wasm_v128_load(table + 192); const v128_t vtD = wasm_v128_load(table + 208); const v128_t vtE = wasm_v128_load(table + 224); const v128_t vtF = wasm_v128_load(table + 240); const v128_t vtable0 = vt0; const v128_t vtable1 = wasm_v128_xor(vt0, vt1); const v128_t vtable2 = wasm_v128_xor(vt1, vt2); const v128_t vtable3 = wasm_v128_xor(vt2, vt3); const v128_t vtable4 = wasm_v128_xor(vt3, vt4); const v128_t vtable5 = wasm_v128_xor(vt4, vt5); const v128_t vtable6 = wasm_v128_xor(vt5, vt6); const v128_t vtable7 = wasm_v128_xor(vt6, vt7); const v128_t vtable8 = wasm_v128_xor(wasm_v128_xor(vt7, vt8), vtable0); const v128_t vtable9 = wasm_v128_xor(wasm_v128_xor(vt8, vt9), vtable1); const v128_t vtableA = wasm_v128_xor(wasm_v128_xor(vt9, vtA), vtable2); const v128_t vtableB = wasm_v128_xor(wasm_v128_xor(vtA, vtB), vtable3); const v128_t vtableC = wasm_v128_xor(wasm_v128_xor(vtB, vtC), vtable4); const v128_t vtableD = wasm_v128_xor(wasm_v128_xor(vtC, vtD), vtable5); const v128_t vtableE = wasm_v128_xor(wasm_v128_xor(vtD, vtE), vtable6); const v128_t vtableF = wasm_v128_xor(wasm_v128_xor(vtE, vtF), vtable7); const v128_t voffset = wasm_i8x16_const_splat(16); for (; batch >= 48 * sizeof(uint8_t); batch -= 48 * sizeof(uint8_t)) { v128_t vx0 = wasm_v128_load((const v128_t*) input); v128_t vx1 = wasm_v128_load((const v128_t*) (input + 16)); v128_t vx2 = wasm_v128_load((const v128_t*) (input + 32)); input += 48; v128_t vy0 = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx0); v128_t vy1 = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx1); v128_t vy2 = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx2); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx2)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx2)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx2)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx2)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx2)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx2)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx2)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx2)); wasm_v128_store(output, vy0); wasm_v128_store(output + 16, vy1); wasm_v128_store(output + 32, vy2); output += 48; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { v128_t vx = wasm_v128_load(input); input += 16; v128_t vy = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx)); wasm_v128_store(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { v128_t vx = wasm_v128_load((const v128_t*) input); v128_t vy = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { wasm_v128_store64_lane(output, vy, 0); vy = wasm_v64x2_shuffle(vy, vy, 1, 1); output += 8; } if (batch & (4 * sizeof(uint8_t))) { wasm_v128_store32_lane(output, vy, 0); vy = wasm_u64x2_shr(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { wasm_v128_store16_lane(output, vy, 0); vy = wasm_u32x4_shr(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { wasm_v128_store8_lane(output, vy, 0); } } }
13,187
48.026022
81
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-wasmpshufb-x64.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/wasmpshufb.c.in // Generator: tools/xngen // // Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <wasm_simd128.h> #include <xnnpack/common.h> #include <xnnpack/lut.h> void xnn_x8_lut_ukernel__wasmpshufb_x64( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const v128_t vt0 = wasm_v128_load(table); const v128_t vt1 = wasm_v128_load(table + 16); const v128_t vt2 = wasm_v128_load(table + 32); const v128_t vt3 = wasm_v128_load(table + 48); const v128_t vt4 = wasm_v128_load(table + 64); const v128_t vt5 = wasm_v128_load(table + 80); const v128_t vt6 = wasm_v128_load(table + 96); const v128_t vt7 = wasm_v128_load(table + 112); const v128_t vt8 = wasm_v128_load(table + 128); const v128_t vt9 = wasm_v128_load(table + 144); const v128_t vtA = wasm_v128_load(table + 160); const v128_t vtB = wasm_v128_load(table + 176); const v128_t vtC = wasm_v128_load(table + 192); const v128_t vtD = wasm_v128_load(table + 208); const v128_t vtE = wasm_v128_load(table + 224); const v128_t vtF = wasm_v128_load(table + 240); const v128_t vtable0 = vt0; const v128_t vtable1 = wasm_v128_xor(vt0, vt1); const v128_t vtable2 = wasm_v128_xor(vt1, vt2); const v128_t vtable3 = wasm_v128_xor(vt2, vt3); const v128_t vtable4 = wasm_v128_xor(vt3, vt4); const v128_t vtable5 = wasm_v128_xor(vt4, vt5); const v128_t vtable6 = wasm_v128_xor(vt5, vt6); const v128_t vtable7 = wasm_v128_xor(vt6, vt7); const v128_t vtable8 = wasm_v128_xor(wasm_v128_xor(vt7, vt8), vtable0); const v128_t vtable9 = wasm_v128_xor(wasm_v128_xor(vt8, vt9), vtable1); const v128_t vtableA = wasm_v128_xor(wasm_v128_xor(vt9, vtA), vtable2); const v128_t vtableB = wasm_v128_xor(wasm_v128_xor(vtA, vtB), vtable3); const v128_t vtableC = wasm_v128_xor(wasm_v128_xor(vtB, vtC), vtable4); const v128_t vtableD = wasm_v128_xor(wasm_v128_xor(vtC, vtD), vtable5); const v128_t vtableE = wasm_v128_xor(wasm_v128_xor(vtD, vtE), vtable6); const v128_t vtableF = wasm_v128_xor(wasm_v128_xor(vtE, vtF), vtable7); const v128_t voffset = wasm_i8x16_const_splat(16); for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { v128_t vx0 = wasm_v128_load((const v128_t*) input); v128_t vx1 = wasm_v128_load((const v128_t*) (input + 16)); v128_t vx2 = wasm_v128_load((const v128_t*) (input + 32)); v128_t vx3 = wasm_v128_load((const v128_t*) (input + 48)); input += 64; v128_t vy0 = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx0); v128_t vy1 = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx1); v128_t vy2 = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx2); v128_t vy3 = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx3); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vx3 = wasm_i8x16_sub(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vx3 = wasm_i8x16_sub(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vx3 = wasm_i8x16_sub(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vx3 = wasm_i8x16_sub(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vx3 = wasm_i8x16_sub(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vx3 = wasm_i8x16_sub(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vx3 = wasm_i8x16_sub(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vx1 = wasm_i8x16_sub(vx1, voffset); vx2 = wasm_i8x16_sub(vx2, voffset); vx3 = wasm_i8x16_sub(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx3)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vx3 = wasm_i8x16_sub_sat(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx3)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vx3 = wasm_i8x16_sub_sat(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx3)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vx3 = wasm_i8x16_sub_sat(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx3)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vx3 = wasm_i8x16_sub_sat(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx3)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vx3 = wasm_i8x16_sub_sat(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx3)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vx3 = wasm_i8x16_sub_sat(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx3)); vx0 = wasm_i8x16_sub_sat(vx0, voffset); vx1 = wasm_i8x16_sub_sat(vx1, voffset); vx2 = wasm_i8x16_sub_sat(vx2, voffset); vx3 = wasm_i8x16_sub_sat(vx3, voffset); vy0 = wasm_v128_xor(vy0, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx0)); vy1 = wasm_v128_xor(vy1, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx1)); vy2 = wasm_v128_xor(vy2, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx2)); vy3 = wasm_v128_xor(vy3, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx3)); wasm_v128_store(output, vy0); wasm_v128_store(output + 16, vy1); wasm_v128_store(output + 32, vy2); wasm_v128_store(output + 48, vy3); output += 64; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { v128_t vx = wasm_v128_load(input); input += 16; v128_t vy = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx)); wasm_v128_store(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { v128_t vx = wasm_v128_load((const v128_t*) input); v128_t vy = __builtin_wasm_relaxed_swizzle_i8x16(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable8, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtable9, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableA, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableB, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableC, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableD, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableE, vx)); vx = wasm_i8x16_sub_sat(vx, voffset); vy = wasm_v128_xor(vy, __builtin_wasm_relaxed_swizzle_i8x16(vtableF, vx)); if (batch & (8 * sizeof(uint8_t))) { wasm_v128_store64_lane(output, vy, 0); vy = wasm_v64x2_shuffle(vy, vy, 1, 1); output += 8; } if (batch & (4 * sizeof(uint8_t))) { wasm_v128_store32_lane(output, vy, 0); vy = wasm_u64x2_shr(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { wasm_v128_store16_lane(output, vy, 0); vy = wasm_u32x4_shr(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { wasm_v128_store8_lane(output, vy, 0); } } }
15,216
49.387417
81
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-wasmsimd-x16.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/wasmsimd.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <wasm_simd128.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__wasmsimd_x16( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const v128_t vtable0 = wasm_v128_load(table); const v128_t vtable1 = wasm_v128_load(table + 16); const v128_t vtable2 = wasm_v128_load(table + 32); const v128_t vtable3 = wasm_v128_load(table + 48); const v128_t vtable4 = wasm_v128_load(table + 64); const v128_t vtable5 = wasm_v128_load(table + 80); const v128_t vtable6 = wasm_v128_load(table + 96); const v128_t vtable7 = wasm_v128_load(table + 112); const v128_t vtable8 = wasm_v128_load(table + 128); const v128_t vtable9 = wasm_v128_load(table + 144); const v128_t vtable10 = wasm_v128_load(table + 160); const v128_t vtable11 = wasm_v128_load(table + 176); const v128_t vtable12 = wasm_v128_load(table + 192); const v128_t vtable13 = wasm_v128_load(table + 208); const v128_t vtable14 = wasm_v128_load(table + 224); const v128_t vtable15 = wasm_v128_load(table + 240); const v128_t voffset = wasm_i8x16_const_splat(16); for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { v128_t vx = wasm_v128_load(input); input += 16; v128_t vy = wasm_i8x16_swizzle(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable8, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable9, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable10, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable11, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable12, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable13, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable14, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable15, vx)); wasm_v128_store(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { v128_t vx = wasm_v128_load(input); v128_t vy = wasm_i8x16_swizzle(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable8, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable9, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable10, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable11, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable12, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable13, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable14, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable15, vx)); if (batch & (8 * sizeof(uint8_t))) { wasm_v128_store64_lane(output, vy, 0); vy = wasm_v64x2_shuffle(vy, vy, 1, 1); output += 8; } if (batch & (4 * sizeof(uint8_t))) { wasm_v128_store32_lane(output, vy, 0); vy = wasm_u64x2_shr(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { wasm_v128_store16_lane(output, vy, 0); vy = wasm_u32x4_shr(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { wasm_v128_store8_lane(output, vy, 0); } } }
5,461
37.195804
72
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-wasmsimd-x32.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/wasmsimd.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <wasm_simd128.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__wasmsimd_x32( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const v128_t vtable0 = wasm_v128_load(table); const v128_t vtable1 = wasm_v128_load(table + 16); const v128_t vtable2 = wasm_v128_load(table + 32); const v128_t vtable3 = wasm_v128_load(table + 48); const v128_t vtable4 = wasm_v128_load(table + 64); const v128_t vtable5 = wasm_v128_load(table + 80); const v128_t vtable6 = wasm_v128_load(table + 96); const v128_t vtable7 = wasm_v128_load(table + 112); const v128_t vtable8 = wasm_v128_load(table + 128); const v128_t vtable9 = wasm_v128_load(table + 144); const v128_t vtable10 = wasm_v128_load(table + 160); const v128_t vtable11 = wasm_v128_load(table + 176); const v128_t vtable12 = wasm_v128_load(table + 192); const v128_t vtable13 = wasm_v128_load(table + 208); const v128_t vtable14 = wasm_v128_load(table + 224); const v128_t vtable15 = wasm_v128_load(table + 240); const v128_t voffset = wasm_i8x16_const_splat(16); for (; batch >= 32 * sizeof(uint8_t); batch -= 32 * sizeof(uint8_t)) { v128_t vx0 = wasm_v128_load(input); v128_t vx1 = wasm_v128_load(input + 16); input += 32; v128_t vy0 = wasm_i8x16_swizzle(vtable0, vx0); v128_t vy1 = wasm_i8x16_swizzle(vtable0, vx1); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable1, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable1, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable2, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable2, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable3, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable3, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable4, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable4, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable5, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable5, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable6, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable6, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable7, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable7, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable8, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable8, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable9, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable9, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable10, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable10, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable11, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable11, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable12, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable12, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable13, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable13, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable14, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable14, vx1)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable15, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable15, vx1)); wasm_v128_store(output, vy0); wasm_v128_store(output + 16, vy1); output += 32; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { v128_t vx = wasm_v128_load(input); input += 16; v128_t vy = wasm_i8x16_swizzle(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable8, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable9, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable10, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable11, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable12, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable13, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable14, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable15, vx)); wasm_v128_store(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { v128_t vx = wasm_v128_load(input); v128_t vy = wasm_i8x16_swizzle(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable8, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable9, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable10, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable11, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable12, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable13, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable14, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable15, vx)); if (batch & (8 * sizeof(uint8_t))) { wasm_v128_store64_lane(output, vy, 0); vy = wasm_v64x2_shuffle(vy, vy, 1, 1); output += 8; } if (batch & (4 * sizeof(uint8_t))) { wasm_v128_store32_lane(output, vy, 0); vy = wasm_u64x2_shr(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { wasm_v128_store16_lane(output, vy, 0); vy = wasm_u32x4_shr(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { wasm_v128_store8_lane(output, vy, 0); } } }
8,938
40.384259
72
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-wasmsimd-x48.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/wasmsimd.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <wasm_simd128.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__wasmsimd_x48( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const v128_t vtable0 = wasm_v128_load(table); const v128_t vtable1 = wasm_v128_load(table + 16); const v128_t vtable2 = wasm_v128_load(table + 32); const v128_t vtable3 = wasm_v128_load(table + 48); const v128_t vtable4 = wasm_v128_load(table + 64); const v128_t vtable5 = wasm_v128_load(table + 80); const v128_t vtable6 = wasm_v128_load(table + 96); const v128_t vtable7 = wasm_v128_load(table + 112); const v128_t vtable8 = wasm_v128_load(table + 128); const v128_t vtable9 = wasm_v128_load(table + 144); const v128_t vtable10 = wasm_v128_load(table + 160); const v128_t vtable11 = wasm_v128_load(table + 176); const v128_t vtable12 = wasm_v128_load(table + 192); const v128_t vtable13 = wasm_v128_load(table + 208); const v128_t vtable14 = wasm_v128_load(table + 224); const v128_t vtable15 = wasm_v128_load(table + 240); const v128_t voffset = wasm_i8x16_const_splat(16); for (; batch >= 48 * sizeof(uint8_t); batch -= 48 * sizeof(uint8_t)) { v128_t vx0 = wasm_v128_load(input); v128_t vx1 = wasm_v128_load(input + 16); v128_t vx2 = wasm_v128_load(input + 32); input += 48; v128_t vy0 = wasm_i8x16_swizzle(vtable0, vx0); v128_t vy1 = wasm_i8x16_swizzle(vtable0, vx1); v128_t vy2 = wasm_i8x16_swizzle(vtable0, vx2); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable1, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable1, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable1, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable2, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable2, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable2, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable3, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable3, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable3, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable4, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable4, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable4, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable5, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable5, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable5, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable6, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable6, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable6, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable7, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable7, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable7, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable8, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable8, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable8, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable9, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable9, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable9, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable10, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable10, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable10, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable11, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable11, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable11, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable12, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable12, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable12, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable13, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable13, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable13, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable14, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable14, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable14, vx2)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable15, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable15, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable15, vx2)); wasm_v128_store(output, vy0); wasm_v128_store(output + 16, vy1); wasm_v128_store(output + 32, vy2); output += 48; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { v128_t vx = wasm_v128_load(input); input += 16; v128_t vy = wasm_i8x16_swizzle(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable8, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable9, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable10, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable11, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable12, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable13, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable14, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable15, vx)); wasm_v128_store(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { v128_t vx = wasm_v128_load(input); v128_t vy = wasm_i8x16_swizzle(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable8, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable9, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable10, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable11, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable12, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable13, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable14, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable15, vx)); if (batch & (8 * sizeof(uint8_t))) { wasm_v128_store64_lane(output, vy, 0); vy = wasm_v64x2_shuffle(vy, vy, 1, 1); output += 8; } if (batch & (4 * sizeof(uint8_t))) { wasm_v128_store32_lane(output, vy, 0); vy = wasm_u64x2_shr(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { wasm_v128_store16_lane(output, vy, 0); vy = wasm_u32x4_shr(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { wasm_v128_store8_lane(output, vy, 0); } } }
10,624
41.670683
72
c
XNNPACK
XNNPACK-master/src/x8-lut/gen/x8-lut-wasmsimd-x64.c
// Auto-generated file. Do not edit! // Template: src/x8-lut/wasmsimd.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <wasm_simd128.h> #include <xnnpack/intrinsics-polyfill.h> #include <xnnpack/lut.h> #include <xnnpack/common.h> void xnn_x8_lut_ukernel__wasmsimd_x64( size_t batch, const uint8_t* input, uint8_t* output, const uint8_t table[restrict XNN_MIN_ELEMENTS(256)]) { assert(batch != 0); assert(batch % sizeof(uint8_t) == 0); assert(input != NULL); assert(output != NULL); const v128_t vtable0 = wasm_v128_load(table); const v128_t vtable1 = wasm_v128_load(table + 16); const v128_t vtable2 = wasm_v128_load(table + 32); const v128_t vtable3 = wasm_v128_load(table + 48); const v128_t vtable4 = wasm_v128_load(table + 64); const v128_t vtable5 = wasm_v128_load(table + 80); const v128_t vtable6 = wasm_v128_load(table + 96); const v128_t vtable7 = wasm_v128_load(table + 112); const v128_t vtable8 = wasm_v128_load(table + 128); const v128_t vtable9 = wasm_v128_load(table + 144); const v128_t vtable10 = wasm_v128_load(table + 160); const v128_t vtable11 = wasm_v128_load(table + 176); const v128_t vtable12 = wasm_v128_load(table + 192); const v128_t vtable13 = wasm_v128_load(table + 208); const v128_t vtable14 = wasm_v128_load(table + 224); const v128_t vtable15 = wasm_v128_load(table + 240); const v128_t voffset = wasm_i8x16_const_splat(16); for (; batch >= 64 * sizeof(uint8_t); batch -= 64 * sizeof(uint8_t)) { v128_t vx0 = wasm_v128_load(input); v128_t vx1 = wasm_v128_load(input + 16); v128_t vx2 = wasm_v128_load(input + 32); v128_t vx3 = wasm_v128_load(input + 48); input += 64; v128_t vy0 = wasm_i8x16_swizzle(vtable0, vx0); v128_t vy1 = wasm_i8x16_swizzle(vtable0, vx1); v128_t vy2 = wasm_i8x16_swizzle(vtable0, vx2); v128_t vy3 = wasm_i8x16_swizzle(vtable0, vx3); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable1, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable1, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable1, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable1, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable2, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable2, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable2, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable2, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable3, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable3, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable3, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable3, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable4, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable4, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable4, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable4, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable5, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable5, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable5, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable5, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable6, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable6, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable6, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable6, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable7, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable7, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable7, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable7, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable8, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable8, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable8, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable8, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable9, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable9, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable9, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable9, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable10, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable10, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable10, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable10, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable11, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable11, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable11, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable11, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable12, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable12, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable12, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable12, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable13, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable13, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable13, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable13, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable14, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable14, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable14, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable14, vx3)); vx0 = wasm_i8x16_sub(vx0, voffset); vy0 = wasm_v128_or(vy0, wasm_i8x16_swizzle(vtable15, vx0)); vx1 = wasm_i8x16_sub(vx1, voffset); vy1 = wasm_v128_or(vy1, wasm_i8x16_swizzle(vtable15, vx1)); vx2 = wasm_i8x16_sub(vx2, voffset); vy2 = wasm_v128_or(vy2, wasm_i8x16_swizzle(vtable15, vx2)); vx3 = wasm_i8x16_sub(vx3, voffset); vy3 = wasm_v128_or(vy3, wasm_i8x16_swizzle(vtable15, vx3)); wasm_v128_store(output, vy0); wasm_v128_store(output + 16, vy1); wasm_v128_store(output + 32, vy2); wasm_v128_store(output + 48, vy3); output += 64; } for (; batch >= 16 * sizeof(uint8_t); batch -= 16 * sizeof(uint8_t)) { v128_t vx = wasm_v128_load(input); input += 16; v128_t vy = wasm_i8x16_swizzle(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable8, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable9, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable10, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable11, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable12, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable13, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable14, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable15, vx)); wasm_v128_store(output, vy); output += 16; } if XNN_UNLIKELY(batch != 0) { v128_t vx = wasm_v128_load(input); v128_t vy = wasm_i8x16_swizzle(vtable0, vx); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable1, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable2, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable3, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable4, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable5, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable6, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable7, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable8, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable9, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable10, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable11, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable12, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable13, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable14, vx)); vx = wasm_i8x16_sub(vx, voffset); vy = wasm_v128_or(vy, wasm_i8x16_swizzle(vtable15, vx)); if (batch & (8 * sizeof(uint8_t))) { wasm_v128_store64_lane(output, vy, 0); vy = wasm_v64x2_shuffle(vy, vy, 1, 1); output += 8; } if (batch & (4 * sizeof(uint8_t))) { wasm_v128_store32_lane(output, vy, 0); vy = wasm_u64x2_shr(vy, 32); output += 4; } if (batch & (2 * sizeof(uint8_t))) { wasm_v128_store16_lane(output, vy, 0); vy = wasm_u32x4_shr(vy, 16); output += 2; } if (batch & (1 * sizeof(uint8_t))) { wasm_v128_store8_lane(output, vy, 0); } } }
12,310
42.656028
72
c
XNNPACK
XNNPACK-master/src/x8-packw/gen/x8-packw-x16-gemm-goi-scalar-int-x4.c
// Auto-generated file. Do not edit! // Template: src/x8-packw/scalar.c.in // Generator: tools/xngen // // Copyright 2023 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <stddef.h> #include <stdint.h> #include <xnnpack/math.h> #include <xnnpack/packw.h> void xnn_x8_packw_gemm_goi_ukernel_x16__scalar_int_x4( size_t g, size_t nc, size_t kc, size_t nr, size_t kr, size_t sr, const int8_t* weights, const uint32_t* bias, int8_t* packed_weights, size_t extra_bytes, const void* params) { assert(g != 0); assert(nc != 0); assert(kc != 0); assert(nr == 16); // This kernel is for NR=16 assert(kr == 1); assert(sr == 1); assert(weights != NULL); assert(packed_weights != NULL); int8_t* out = (int8_t*) packed_weights; const uint32_t* b = (const uint32_t*) bias; do { // NC main loop multiple of 16 const int8_t* w0 = (const int8_t*) weights; size_t n = nc; for (;n >= 16; n -= 16) { if XNN_LIKELY(b != NULL) { ((uint32_t*) out)[0] = b[0]; ((uint32_t*) out)[1] = b[1]; ((uint32_t*) out)[2] = b[2]; ((uint32_t*) out)[3] = b[3]; ((uint32_t*) out)[4] = b[4]; ((uint32_t*) out)[5] = b[5]; ((uint32_t*) out)[6] = b[6]; ((uint32_t*) out)[7] = b[7]; ((uint32_t*) out)[8] = b[8]; ((uint32_t*) out)[9] = b[9]; ((uint32_t*) out)[10] = b[10]; ((uint32_t*) out)[11] = b[11]; ((uint32_t*) out)[12] = b[12]; ((uint32_t*) out)[13] = b[13]; ((uint32_t*) out)[14] = b[14]; ((uint32_t*) out)[15] = b[15]; b += 16; } else { ((uint32_t*) out)[0] = 0; ((uint32_t*) out)[1] = 0; ((uint32_t*) out)[2] = 0; ((uint32_t*) out)[3] = 0; ((uint32_t*) out)[4] = 0; ((uint32_t*) out)[5] = 0; ((uint32_t*) out)[6] = 0; ((uint32_t*) out)[7] = 0; ((uint32_t*) out)[8] = 0; ((uint32_t*) out)[9] = 0; ((uint32_t*) out)[10] = 0; ((uint32_t*) out)[11] = 0; ((uint32_t*) out)[12] = 0; ((uint32_t*) out)[13] = 0; ((uint32_t*) out)[14] = 0; ((uint32_t*) out)[15] = 0; } out += 16 * sizeof(uint32_t); const int8_t* w1 = w0 + kc; const int8_t* w2 = w1 + kc; const int8_t* w3 = w2 + kc; const int8_t* w4 = w3 + kc; const int8_t* w5 = w4 + kc; const int8_t* w6 = w5 + kc; const int8_t* w7 = w6 + kc; const int8_t* w8 = w7 + kc; const int8_t* w9 = w8 + kc; const int8_t* w10 = w9 + kc; const int8_t* w11 = w10 + kc; const int8_t* w12 = w11 + kc; const int8_t* w13 = w12 + kc; const int8_t* w14 = w13 + kc; const int8_t* w15 = w14 + kc; // KC main loop multiple of 16x4 size_t k = kc; for (; k >= 4; k -= 4) { const int8_t v00 = w0[0]; const int8_t v01 = w0[1]; const int8_t v02 = w0[2]; const int8_t v03 = w0[3]; w0 += 4; const int8_t v10 = w1[0]; const int8_t v11 = w1[1]; const int8_t v12 = w1[2]; const int8_t v13 = w1[3]; w1 += 4; const int8_t v20 = w2[0]; const int8_t v21 = w2[1]; const int8_t v22 = w2[2]; const int8_t v23 = w2[3]; w2 += 4; const int8_t v30 = w3[0]; const int8_t v31 = w3[1]; const int8_t v32 = w3[2]; const int8_t v33 = w3[3]; w3 += 4; const int8_t v40 = w4[0]; const int8_t v41 = w4[1]; const int8_t v42 = w4[2]; const int8_t v43 = w4[3]; w4 += 4; const int8_t v50 = w5[0]; const int8_t v51 = w5[1]; const int8_t v52 = w5[2]; const int8_t v53 = w5[3]; w5 += 4; const int8_t v60 = w6[0]; const int8_t v61 = w6[1]; const int8_t v62 = w6[2]; const int8_t v63 = w6[3]; w6 += 4; const int8_t v70 = w7[0]; const int8_t v71 = w7[1]; const int8_t v72 = w7[2]; const int8_t v73 = w7[3]; w7 += 4; const int8_t v80 = w8[0]; const int8_t v81 = w8[1]; const int8_t v82 = w8[2]; const int8_t v83 = w8[3]; w8 += 4; const int8_t v90 = w9[0]; const int8_t v91 = w9[1]; const int8_t v92 = w9[2]; const int8_t v93 = w9[3]; w9 += 4; const int8_t v100 = w10[0]; const int8_t v101 = w10[1]; const int8_t v102 = w10[2]; const int8_t v103 = w10[3]; w10 += 4; const int8_t v110 = w11[0]; const int8_t v111 = w11[1]; const int8_t v112 = w11[2]; const int8_t v113 = w11[3]; w11 += 4; const int8_t v120 = w12[0]; const int8_t v121 = w12[1]; const int8_t v122 = w12[2]; const int8_t v123 = w12[3]; w12 += 4; const int8_t v130 = w13[0]; const int8_t v131 = w13[1]; const int8_t v132 = w13[2]; const int8_t v133 = w13[3]; w13 += 4; const int8_t v140 = w14[0]; const int8_t v141 = w14[1]; const int8_t v142 = w14[2]; const int8_t v143 = w14[3]; w14 += 4; const int8_t v150 = w15[0]; const int8_t v151 = w15[1]; const int8_t v152 = w15[2]; const int8_t v153 = w15[3]; w15 += 4; out[0] = v00; out[1] = v10; out[2] = v20; out[3] = v30; out[4] = v40; out[5] = v50; out[6] = v60; out[7] = v70; out[8] = v80; out[9] = v90; out[10] = v100; out[11] = v110; out[12] = v120; out[13] = v130; out[14] = v140; out[15] = v150; out[16] = v01; out[17] = v11; out[18] = v21; out[19] = v31; out[20] = v41; out[21] = v51; out[22] = v61; out[23] = v71; out[24] = v81; out[25] = v91; out[26] = v101; out[27] = v111; out[28] = v121; out[29] = v131; out[30] = v141; out[31] = v151; out[32] = v02; out[33] = v12; out[34] = v22; out[35] = v32; out[36] = v42; out[37] = v52; out[38] = v62; out[39] = v72; out[40] = v82; out[41] = v92; out[42] = v102; out[43] = v112; out[44] = v122; out[45] = v132; out[46] = v142; out[47] = v152; out[48] = v03; out[49] = v13; out[50] = v23; out[51] = v33; out[52] = v43; out[53] = v53; out[54] = v63; out[55] = v73; out[56] = v83; out[57] = v93; out[58] = v103; out[59] = v113; out[60] = v123; out[61] = v133; out[62] = v143; out[63] = v153; out += 64; } // KC remainder for (; k != 0; --k) { const int8_t v0 = *w0++; out[0] = v0; const int8_t v1 = *w1++; out[1] = v1; const int8_t v2 = *w2++; out[2] = v2; const int8_t v3 = *w3++; out[3] = v3; const int8_t v4 = *w4++; out[4] = v4; const int8_t v5 = *w5++; out[5] = v5; const int8_t v6 = *w6++; out[6] = v6; const int8_t v7 = *w7++; out[7] = v7; const int8_t v8 = *w8++; out[8] = v8; const int8_t v9 = *w9++; out[9] = v9; const int8_t v10 = *w10++; out[10] = v10; const int8_t v11 = *w11++; out[11] = v11; const int8_t v12 = *w12++; out[12] = v12; const int8_t v13 = *w13++; out[13] = v13; const int8_t v14 = *w14++; out[14] = v14; const int8_t v15 = *w15++; out[15] = v15; out += 16; } out = (int8_t*) ((uintptr_t) out + extra_bytes); w0 = w15; } // NC remainder (1..15) if XNN_UNLIKELY(n != 0) { if XNN_LIKELY(b != NULL) { size_t nb = n; do { *((uint32_t*) out) = *b++; out += sizeof(uint32_t); } while (--nb != 0); } else { size_t nb = n; do { *((uint32_t*) out) = 0; out += sizeof(uint32_t); } while (--nb != 0); } out += (16 - n) * sizeof(uint32_t); // NR remainder has less than 16 rows so last row is not loaded const int8_t* w1 = w0 + kc; if XNN_UNPREDICTABLE(n < 2) { w1 = w0; } const int8_t* w2 = w1 + kc; if XNN_UNPREDICTABLE(n <= 2) { w2 = w1; } const int8_t* w3 = w2 + kc; if XNN_UNPREDICTABLE(n < 4) { w3 = w2; } const int8_t* w4 = w3 + kc; if XNN_UNPREDICTABLE(n <= 4) { w4 = w3; } const int8_t* w5 = w4 + kc; if XNN_UNPREDICTABLE(n < 6) { w5 = w4; } const int8_t* w6 = w5 + kc; if XNN_UNPREDICTABLE(n <= 6) { w6 = w5; } const int8_t* w7 = w6 + kc; if XNN_UNPREDICTABLE(n < 8) { w7 = w6; } const int8_t* w8 = w7 + kc; if XNN_UNPREDICTABLE(n <= 8) { w8 = w7; } const int8_t* w9 = w8 + kc; if XNN_UNPREDICTABLE(n < 10) { w9 = w8; } const int8_t* w10 = w9 + kc; if XNN_UNPREDICTABLE(n <= 10) { w10 = w9; } const int8_t* w11 = w10 + kc; if XNN_UNPREDICTABLE(n < 12) { w11 = w10; } const int8_t* w12 = w11 + kc; if XNN_UNPREDICTABLE(n <= 12) { w12 = w11; } const int8_t* w13 = w12 + kc; if XNN_UNPREDICTABLE(n < 14) { w13 = w12; } const int8_t* w14 = w13 + kc; if XNN_UNPREDICTABLE(n <= 14) { w14 = w13; } // KC main loop multiple of 16x4 size_t k = kc; for (; k >= 4; k -= 4) { const int8_t v00 = w0[0]; const int8_t v01 = w0[1]; const int8_t v02 = w0[2]; const int8_t v03 = w0[3]; w0 += 4; const int8_t v10 = w1[0]; const int8_t v11 = w1[1]; const int8_t v12 = w1[2]; const int8_t v13 = w1[3]; w1 += 4; const int8_t v20 = w2[0]; const int8_t v21 = w2[1]; const int8_t v22 = w2[2]; const int8_t v23 = w2[3]; w2 += 4; const int8_t v30 = w3[0]; const int8_t v31 = w3[1]; const int8_t v32 = w3[2]; const int8_t v33 = w3[3]; w3 += 4; const int8_t v40 = w4[0]; const int8_t v41 = w4[1]; const int8_t v42 = w4[2]; const int8_t v43 = w4[3]; w4 += 4; const int8_t v50 = w5[0]; const int8_t v51 = w5[1]; const int8_t v52 = w5[2]; const int8_t v53 = w5[3]; w5 += 4; const int8_t v60 = w6[0]; const int8_t v61 = w6[1]; const int8_t v62 = w6[2]; const int8_t v63 = w6[3]; w6 += 4; const int8_t v70 = w7[0]; const int8_t v71 = w7[1]; const int8_t v72 = w7[2]; const int8_t v73 = w7[3]; w7 += 4; const int8_t v80 = w8[0]; const int8_t v81 = w8[1]; const int8_t v82 = w8[2]; const int8_t v83 = w8[3]; w8 += 4; const int8_t v90 = w9[0]; const int8_t v91 = w9[1]; const int8_t v92 = w9[2]; const int8_t v93 = w9[3]; w9 += 4; const int8_t v100 = w10[0]; const int8_t v101 = w10[1]; const int8_t v102 = w10[2]; const int8_t v103 = w10[3]; w10 += 4; const int8_t v110 = w11[0]; const int8_t v111 = w11[1]; const int8_t v112 = w11[2]; const int8_t v113 = w11[3]; w11 += 4; const int8_t v120 = w12[0]; const int8_t v121 = w12[1]; const int8_t v122 = w12[2]; const int8_t v123 = w12[3]; w12 += 4; const int8_t v130 = w13[0]; const int8_t v131 = w13[1]; const int8_t v132 = w13[2]; const int8_t v133 = w13[3]; w13 += 4; const int8_t v140 = w14[0]; const int8_t v141 = w14[1]; const int8_t v142 = w14[2]; const int8_t v143 = w14[3]; w14 += 4; out[0] = v00; out[1] = v10; out[2] = v20; out[3] = v30; out[4] = v40; out[5] = v50; out[6] = v60; out[7] = v70; out[8] = v80; out[9] = v90; out[10] = v100; out[11] = v110; out[12] = v120; out[13] = v130; out[14] = v140; out[16] = v01; out[17] = v11; out[18] = v21; out[19] = v31; out[20] = v41; out[21] = v51; out[22] = v61; out[23] = v71; out[24] = v81; out[25] = v91; out[26] = v101; out[27] = v111; out[28] = v121; out[29] = v131; out[30] = v141; out[32] = v02; out[33] = v12; out[34] = v22; out[35] = v32; out[36] = v42; out[37] = v52; out[38] = v62; out[39] = v72; out[40] = v82; out[41] = v92; out[42] = v102; out[43] = v112; out[44] = v122; out[45] = v132; out[46] = v142; out[48] = v03; out[49] = v13; out[50] = v23; out[51] = v33; out[52] = v43; out[53] = v53; out[54] = v63; out[55] = v73; out[56] = v83; out[57] = v93; out[58] = v103; out[59] = v113; out[60] = v123; out[61] = v133; out[62] = v143; out += 64; } // KC remainder of 1..3 for (; k != 0; --k) { const int8_t v0 = *w0++; out[0] = v0; const int8_t v1 = *w1++; out[1] = v1; const int8_t v2 = *w2++; out[2] = v2; const int8_t v3 = *w3++; out[3] = v3; const int8_t v4 = *w4++; out[4] = v4; const int8_t v5 = *w5++; out[5] = v5; const int8_t v6 = *w6++; out[6] = v6; const int8_t v7 = *w7++; out[7] = v7; const int8_t v8 = *w8++; out[8] = v8; const int8_t v9 = *w9++; out[9] = v9; const int8_t v10 = *w10++; out[10] = v10; const int8_t v11 = *w11++; out[11] = v11; const int8_t v12 = *w12++; out[12] = v12; const int8_t v13 = *w13++; out[13] = v13; const int8_t v14 = *w14++; out[14] = v14; out += 16; } out = (int8_t*) ((uintptr_t) out + extra_bytes); } weights += nc * kc; } while (--g != 0); }
14,747
25.863388
72
c
XNNPACK
XNNPACK-master/src/x8-packw/gen/x8-packw-x2-gemm-goi-scalar-int-x4.c
// Auto-generated file. Do not edit! // Template: src/x8-packw/scalar.c.in // Generator: tools/xngen // // Copyright 2023 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <stddef.h> #include <stdint.h> #include <xnnpack/math.h> #include <xnnpack/packw.h> void xnn_x8_packw_gemm_goi_ukernel_x2__scalar_int_x4( size_t g, size_t nc, size_t kc, size_t nr, size_t kr, size_t sr, const int8_t* weights, const uint32_t* bias, int8_t* packed_weights, size_t extra_bytes, const void* params) { assert(g != 0); assert(nc != 0); assert(kc != 0); assert(nr == 2); // This kernel is for NR=2 assert(kr == 1); assert(sr == 1); assert(weights != NULL); assert(packed_weights != NULL); int8_t* out = (int8_t*) packed_weights; const uint32_t* b = (const uint32_t*) bias; do { // NC main loop multiple of 2 const int8_t* w0 = (const int8_t*) weights; size_t n = nc; for (;n >= 2; n -= 2) { if XNN_LIKELY(b != NULL) { ((uint32_t*) out)[0] = b[0]; ((uint32_t*) out)[1] = b[1]; b += 2; } else { ((uint32_t*) out)[0] = 0; ((uint32_t*) out)[1] = 0; } out += 2 * sizeof(uint32_t); const int8_t* w1 = w0 + kc; // KC main loop multiple of 2x4 size_t k = kc; for (; k >= 4; k -= 4) { const int8_t v00 = w0[0]; const int8_t v01 = w0[1]; const int8_t v02 = w0[2]; const int8_t v03 = w0[3]; w0 += 4; const int8_t v10 = w1[0]; const int8_t v11 = w1[1]; const int8_t v12 = w1[2]; const int8_t v13 = w1[3]; w1 += 4; out[0] = v00; out[1] = v10; out[2] = v01; out[3] = v11; out[4] = v02; out[5] = v12; out[6] = v03; out[7] = v13; out += 8; } // KC remainder for (; k != 0; --k) { const int8_t v0 = *w0++; out[0] = v0; const int8_t v1 = *w1++; out[1] = v1; out += 2; } out = (int8_t*) ((uintptr_t) out + extra_bytes); w0 = w1; } // NC remainder (1..1) if XNN_UNLIKELY(n != 0) { if XNN_LIKELY(b != NULL) { size_t nb = n; do { *((uint32_t*) out) = *b++; out += sizeof(uint32_t); } while (--nb != 0); } else { size_t nb = n; do { *((uint32_t*) out) = 0; out += sizeof(uint32_t); } while (--nb != 0); } out += (2 - n) * sizeof(uint32_t); // KC main loop multiple of 2x4 size_t k = kc; for (; k >= 4; k -= 4) { const int8_t v00 = w0[0]; const int8_t v01 = w0[1]; const int8_t v02 = w0[2]; const int8_t v03 = w0[3]; w0 += 4; out[0] = v00; out[2] = v01; out[4] = v02; out[6] = v03; out += 8; } // KC remainder of 1..3 for (; k != 0; --k) { const int8_t v0 = *w0++; out[0] = v0; out += 2; } out = (int8_t*) ((uintptr_t) out + extra_bytes); } weights += nc * kc; } while (--g != 0); }
3,249
21.887324
72
c
XNNPACK
XNNPACK-master/src/x8-packw/gen/x8-packw-x4-gemm-goi-scalar-int-x4.c
// Auto-generated file. Do not edit! // Template: src/x8-packw/scalar.c.in // Generator: tools/xngen // // Copyright 2023 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <stddef.h> #include <stdint.h> #include <xnnpack/math.h> #include <xnnpack/packw.h> void xnn_x8_packw_gemm_goi_ukernel_x4__scalar_int_x4( size_t g, size_t nc, size_t kc, size_t nr, size_t kr, size_t sr, const int8_t* weights, const uint32_t* bias, int8_t* packed_weights, size_t extra_bytes, const void* params) { assert(g != 0); assert(nc != 0); assert(kc != 0); assert(nr == 4); // This kernel is for NR=4 assert(kr == 1); assert(sr == 1); assert(weights != NULL); assert(packed_weights != NULL); int8_t* out = (int8_t*) packed_weights; const uint32_t* b = (const uint32_t*) bias; do { // NC main loop multiple of 4 const int8_t* w0 = (const int8_t*) weights; size_t n = nc; for (;n >= 4; n -= 4) { if XNN_LIKELY(b != NULL) { ((uint32_t*) out)[0] = b[0]; ((uint32_t*) out)[1] = b[1]; ((uint32_t*) out)[2] = b[2]; ((uint32_t*) out)[3] = b[3]; b += 4; } else { ((uint32_t*) out)[0] = 0; ((uint32_t*) out)[1] = 0; ((uint32_t*) out)[2] = 0; ((uint32_t*) out)[3] = 0; } out += 4 * sizeof(uint32_t); const int8_t* w1 = w0 + kc; const int8_t* w2 = w1 + kc; const int8_t* w3 = w2 + kc; // KC main loop multiple of 4x4 size_t k = kc; for (; k >= 4; k -= 4) { const int8_t v00 = w0[0]; const int8_t v01 = w0[1]; const int8_t v02 = w0[2]; const int8_t v03 = w0[3]; w0 += 4; const int8_t v10 = w1[0]; const int8_t v11 = w1[1]; const int8_t v12 = w1[2]; const int8_t v13 = w1[3]; w1 += 4; const int8_t v20 = w2[0]; const int8_t v21 = w2[1]; const int8_t v22 = w2[2]; const int8_t v23 = w2[3]; w2 += 4; const int8_t v30 = w3[0]; const int8_t v31 = w3[1]; const int8_t v32 = w3[2]; const int8_t v33 = w3[3]; w3 += 4; out[0] = v00; out[1] = v10; out[2] = v20; out[3] = v30; out[4] = v01; out[5] = v11; out[6] = v21; out[7] = v31; out[8] = v02; out[9] = v12; out[10] = v22; out[11] = v32; out[12] = v03; out[13] = v13; out[14] = v23; out[15] = v33; out += 16; } // KC remainder for (; k != 0; --k) { const int8_t v0 = *w0++; out[0] = v0; const int8_t v1 = *w1++; out[1] = v1; const int8_t v2 = *w2++; out[2] = v2; const int8_t v3 = *w3++; out[3] = v3; out += 4; } out = (int8_t*) ((uintptr_t) out + extra_bytes); w0 = w3; } // NC remainder (1..3) if XNN_UNLIKELY(n != 0) { if XNN_LIKELY(b != NULL) { size_t nb = n; do { *((uint32_t*) out) = *b++; out += sizeof(uint32_t); } while (--nb != 0); } else { size_t nb = n; do { *((uint32_t*) out) = 0; out += sizeof(uint32_t); } while (--nb != 0); } out += (4 - n) * sizeof(uint32_t); // NR remainder has less than 4 rows so last row is not loaded const int8_t* w1 = w0 + kc; if XNN_UNPREDICTABLE(n < 2) { w1 = w0; } const int8_t* w2 = w1 + kc; if XNN_UNPREDICTABLE(n <= 2) { w2 = w1; } // KC main loop multiple of 4x4 size_t k = kc; for (; k >= 4; k -= 4) { const int8_t v00 = w0[0]; const int8_t v01 = w0[1]; const int8_t v02 = w0[2]; const int8_t v03 = w0[3]; w0 += 4; const int8_t v10 = w1[0]; const int8_t v11 = w1[1]; const int8_t v12 = w1[2]; const int8_t v13 = w1[3]; w1 += 4; const int8_t v20 = w2[0]; const int8_t v21 = w2[1]; const int8_t v22 = w2[2]; const int8_t v23 = w2[3]; w2 += 4; out[0] = v00; out[1] = v10; out[2] = v20; out[4] = v01; out[5] = v11; out[6] = v21; out[8] = v02; out[9] = v12; out[10] = v22; out[12] = v03; out[13] = v13; out[14] = v23; out += 16; } // KC remainder of 1..3 for (; k != 0; --k) { const int8_t v0 = *w0++; out[0] = v0; const int8_t v1 = *w1++; out[1] = v1; const int8_t v2 = *w2++; out[2] = v2; out += 4; } out = (int8_t*) ((uintptr_t) out + extra_bytes); } weights += nc * kc; } while (--g != 0); }
4,911
23.437811
72
c
XNNPACK
XNNPACK-master/src/x8-packw/gen/x8-packw-x8-gemm-goi-scalar-int-x4.c
// Auto-generated file. Do not edit! // Template: src/x8-packw/scalar.c.in // Generator: tools/xngen // // Copyright 2023 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <stddef.h> #include <stdint.h> #include <xnnpack/math.h> #include <xnnpack/packw.h> void xnn_x8_packw_gemm_goi_ukernel_x8__scalar_int_x4( size_t g, size_t nc, size_t kc, size_t nr, size_t kr, size_t sr, const int8_t* weights, const uint32_t* bias, int8_t* packed_weights, size_t extra_bytes, const void* params) { assert(g != 0); assert(nc != 0); assert(kc != 0); assert(nr == 8); // This kernel is for NR=8 assert(kr == 1); assert(sr == 1); assert(weights != NULL); assert(packed_weights != NULL); int8_t* out = (int8_t*) packed_weights; const uint32_t* b = (const uint32_t*) bias; do { // NC main loop multiple of 8 const int8_t* w0 = (const int8_t*) weights; size_t n = nc; for (;n >= 8; n -= 8) { if XNN_LIKELY(b != NULL) { ((uint32_t*) out)[0] = b[0]; ((uint32_t*) out)[1] = b[1]; ((uint32_t*) out)[2] = b[2]; ((uint32_t*) out)[3] = b[3]; ((uint32_t*) out)[4] = b[4]; ((uint32_t*) out)[5] = b[5]; ((uint32_t*) out)[6] = b[6]; ((uint32_t*) out)[7] = b[7]; b += 8; } else { ((uint32_t*) out)[0] = 0; ((uint32_t*) out)[1] = 0; ((uint32_t*) out)[2] = 0; ((uint32_t*) out)[3] = 0; ((uint32_t*) out)[4] = 0; ((uint32_t*) out)[5] = 0; ((uint32_t*) out)[6] = 0; ((uint32_t*) out)[7] = 0; } out += 8 * sizeof(uint32_t); const int8_t* w1 = w0 + kc; const int8_t* w2 = w1 + kc; const int8_t* w3 = w2 + kc; const int8_t* w4 = w3 + kc; const int8_t* w5 = w4 + kc; const int8_t* w6 = w5 + kc; const int8_t* w7 = w6 + kc; // KC main loop multiple of 8x4 size_t k = kc; for (; k >= 4; k -= 4) { const int8_t v00 = w0[0]; const int8_t v01 = w0[1]; const int8_t v02 = w0[2]; const int8_t v03 = w0[3]; w0 += 4; const int8_t v10 = w1[0]; const int8_t v11 = w1[1]; const int8_t v12 = w1[2]; const int8_t v13 = w1[3]; w1 += 4; const int8_t v20 = w2[0]; const int8_t v21 = w2[1]; const int8_t v22 = w2[2]; const int8_t v23 = w2[3]; w2 += 4; const int8_t v30 = w3[0]; const int8_t v31 = w3[1]; const int8_t v32 = w3[2]; const int8_t v33 = w3[3]; w3 += 4; const int8_t v40 = w4[0]; const int8_t v41 = w4[1]; const int8_t v42 = w4[2]; const int8_t v43 = w4[3]; w4 += 4; const int8_t v50 = w5[0]; const int8_t v51 = w5[1]; const int8_t v52 = w5[2]; const int8_t v53 = w5[3]; w5 += 4; const int8_t v60 = w6[0]; const int8_t v61 = w6[1]; const int8_t v62 = w6[2]; const int8_t v63 = w6[3]; w6 += 4; const int8_t v70 = w7[0]; const int8_t v71 = w7[1]; const int8_t v72 = w7[2]; const int8_t v73 = w7[3]; w7 += 4; out[0] = v00; out[1] = v10; out[2] = v20; out[3] = v30; out[4] = v40; out[5] = v50; out[6] = v60; out[7] = v70; out[8] = v01; out[9] = v11; out[10] = v21; out[11] = v31; out[12] = v41; out[13] = v51; out[14] = v61; out[15] = v71; out[16] = v02; out[17] = v12; out[18] = v22; out[19] = v32; out[20] = v42; out[21] = v52; out[22] = v62; out[23] = v72; out[24] = v03; out[25] = v13; out[26] = v23; out[27] = v33; out[28] = v43; out[29] = v53; out[30] = v63; out[31] = v73; out += 32; } // KC remainder for (; k != 0; --k) { const int8_t v0 = *w0++; out[0] = v0; const int8_t v1 = *w1++; out[1] = v1; const int8_t v2 = *w2++; out[2] = v2; const int8_t v3 = *w3++; out[3] = v3; const int8_t v4 = *w4++; out[4] = v4; const int8_t v5 = *w5++; out[5] = v5; const int8_t v6 = *w6++; out[6] = v6; const int8_t v7 = *w7++; out[7] = v7; out += 8; } out = (int8_t*) ((uintptr_t) out + extra_bytes); w0 = w7; } // NC remainder (1..7) if XNN_UNLIKELY(n != 0) { if XNN_LIKELY(b != NULL) { size_t nb = n; do { *((uint32_t*) out) = *b++; out += sizeof(uint32_t); } while (--nb != 0); } else { size_t nb = n; do { *((uint32_t*) out) = 0; out += sizeof(uint32_t); } while (--nb != 0); } out += (8 - n) * sizeof(uint32_t); // NR remainder has less than 8 rows so last row is not loaded const int8_t* w1 = w0 + kc; if XNN_UNPREDICTABLE(n < 2) { w1 = w0; } const int8_t* w2 = w1 + kc; if XNN_UNPREDICTABLE(n <= 2) { w2 = w1; } const int8_t* w3 = w2 + kc; if XNN_UNPREDICTABLE(n < 4) { w3 = w2; } const int8_t* w4 = w3 + kc; if XNN_UNPREDICTABLE(n <= 4) { w4 = w3; } const int8_t* w5 = w4 + kc; if XNN_UNPREDICTABLE(n < 6) { w5 = w4; } const int8_t* w6 = w5 + kc; if XNN_UNPREDICTABLE(n <= 6) { w6 = w5; } // KC main loop multiple of 8x4 size_t k = kc; for (; k >= 4; k -= 4) { const int8_t v00 = w0[0]; const int8_t v01 = w0[1]; const int8_t v02 = w0[2]; const int8_t v03 = w0[3]; w0 += 4; const int8_t v10 = w1[0]; const int8_t v11 = w1[1]; const int8_t v12 = w1[2]; const int8_t v13 = w1[3]; w1 += 4; const int8_t v20 = w2[0]; const int8_t v21 = w2[1]; const int8_t v22 = w2[2]; const int8_t v23 = w2[3]; w2 += 4; const int8_t v30 = w3[0]; const int8_t v31 = w3[1]; const int8_t v32 = w3[2]; const int8_t v33 = w3[3]; w3 += 4; const int8_t v40 = w4[0]; const int8_t v41 = w4[1]; const int8_t v42 = w4[2]; const int8_t v43 = w4[3]; w4 += 4; const int8_t v50 = w5[0]; const int8_t v51 = w5[1]; const int8_t v52 = w5[2]; const int8_t v53 = w5[3]; w5 += 4; const int8_t v60 = w6[0]; const int8_t v61 = w6[1]; const int8_t v62 = w6[2]; const int8_t v63 = w6[3]; w6 += 4; out[0] = v00; out[1] = v10; out[2] = v20; out[3] = v30; out[4] = v40; out[5] = v50; out[6] = v60; out[8] = v01; out[9] = v11; out[10] = v21; out[11] = v31; out[12] = v41; out[13] = v51; out[14] = v61; out[16] = v02; out[17] = v12; out[18] = v22; out[19] = v32; out[20] = v42; out[21] = v52; out[22] = v62; out[24] = v03; out[25] = v13; out[26] = v23; out[27] = v33; out[28] = v43; out[29] = v53; out[30] = v63; out += 32; } // KC remainder of 1..3 for (; k != 0; --k) { const int8_t v0 = *w0++; out[0] = v0; const int8_t v1 = *w1++; out[1] = v1; const int8_t v2 = *w2++; out[2] = v2; const int8_t v3 = *w3++; out[3] = v3; const int8_t v4 = *w4++; out[4] = v4; const int8_t v5 = *w5++; out[5] = v5; const int8_t v6 = *w6++; out[6] = v6; out += 8; } out = (int8_t*) ((uintptr_t) out + extra_bytes); } weights += nc * kc; } while (--g != 0); }
8,104
24.567823
72
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-16x16-reuse-switch-zip-neon.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/neon-zip.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__16x16_reuse_switch_zip_neon( const uint8_t* input, uint8_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(uint8_t)); assert(input_stride >= block_width * sizeof(uint8_t)); const size_t tile_height = 16; const size_t tile_width = 16; const size_t tile_hbytes = tile_height * sizeof(uint8_t); const size_t tile_wbytes = tile_width * sizeof(uint8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(uint8_t); const uint8_t* i0 = input; uint8_t* o = (uint8_t*) output; const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 15); const size_t oN_stride = rem * output_stride; size_t bh = block_height; for (; bh >= 16; bh -= 16) { const uint8x16_t v4_0 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_1 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_2 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_3 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_4 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_5 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_6 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_7 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_8 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_9 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_10 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_11 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_12 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_13 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_14 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16_t v4_15 = vld1q_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x16x2_t v3_0 = vzipq_u8(v4_0, v4_8); const uint8x16x2_t v3_1 = vzipq_u8(v4_1, v4_9); const uint8x16x2_t v3_2 = vzipq_u8(v4_2, v4_10); const uint8x16x2_t v3_3 = vzipq_u8(v4_3, v4_11); const uint8x16x2_t v3_4 = vzipq_u8(v4_4, v4_12); const uint8x16x2_t v3_5 = vzipq_u8(v4_5, v4_13); const uint8x16x2_t v3_6 = vzipq_u8(v4_6, v4_14); const uint8x16x2_t v3_7 = vzipq_u8(v4_7, v4_15); const uint8x16x2_t v2_0 = vzipq_u8(v3_0.val[0], v3_4.val[0]); const uint8x16x2_t v2_1 = vzipq_u8(v3_0.val[1], v3_4.val[1]); const uint8x16x2_t v2_2 = vzipq_u8(v3_1.val[0], v3_5.val[0]); const uint8x16x2_t v2_3 = vzipq_u8(v3_1.val[1], v3_5.val[1]); const uint8x16x2_t v2_4 = vzipq_u8(v3_2.val[0], v3_6.val[0]); const uint8x16x2_t v2_5 = vzipq_u8(v3_2.val[1], v3_6.val[1]); const uint8x16x2_t v2_6 = vzipq_u8(v3_3.val[0], v3_7.val[0]); const uint8x16x2_t v2_7 = vzipq_u8(v3_3.val[1], v3_7.val[1]); const uint8x16x2_t v1_0 = vzipq_u8(v2_0.val[0], v2_4.val[0]); const uint8x16x2_t v1_1 = vzipq_u8(v2_0.val[1], v2_4.val[1]); const uint8x16x2_t v1_2 = vzipq_u8(v2_1.val[0], v2_5.val[0]); const uint8x16x2_t v1_3 = vzipq_u8(v2_1.val[1], v2_5.val[1]); const uint8x16x2_t v1_4 = vzipq_u8(v2_2.val[0], v2_6.val[0]); const uint8x16x2_t v1_5 = vzipq_u8(v2_2.val[1], v2_6.val[1]); const uint8x16x2_t v1_6 = vzipq_u8(v2_3.val[0], v2_7.val[0]); const uint8x16x2_t v1_7 = vzipq_u8(v2_3.val[1], v2_7.val[1]); const uint8x16x2_t v0_0 = vzipq_u8(v1_0.val[0], v1_4.val[0]); const uint8x16x2_t v0_1 = vzipq_u8(v1_0.val[1], v1_4.val[1]); const uint8x16x2_t v0_2 = vzipq_u8(v1_1.val[0], v1_5.val[0]); const uint8x16x2_t v0_3 = vzipq_u8(v1_1.val[1], v1_5.val[1]); const uint8x16x2_t v0_4 = vzipq_u8(v1_2.val[0], v1_6.val[0]); const uint8x16x2_t v0_5 = vzipq_u8(v1_2.val[1], v1_6.val[1]); const uint8x16x2_t v0_6 = vzipq_u8(v1_3.val[0], v1_7.val[0]); const uint8x16x2_t v0_7 = vzipq_u8(v1_3.val[1], v1_7.val[1]); uint8_t *oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 15: vst1q_u8(oN, v0_7.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 14: vst1q_u8(oN, v0_7.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 13: vst1q_u8(oN, v0_6.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 12: vst1q_u8(oN, v0_6.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 11: vst1q_u8(oN, v0_5.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 10: vst1q_u8(oN, v0_5.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 9: vst1q_u8(oN, v0_4.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 8: vst1q_u8(oN, v0_4.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 7: vst1q_u8(oN, v0_3.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1q_u8(oN, v0_3.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1q_u8(oN, v0_2.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1q_u8(oN, v0_2.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1q_u8(oN, v0_1.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1q_u8(oN, v0_1.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1q_u8(oN, v0_0.val[1]); case 0: vst1q_u8(o, v0_0.val[0]); o = (uint8_t*) ((uintptr_t) o + tile_hbytes); break; default: XNN_UNREACHABLE; } } if (bh != 0) { const uint8x16_t v4_0 = vld1q_u8(i0); const uint8_t *i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const uint8x16_t v4_1 = vld1q_u8(i1); const uint8_t *i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i1; } const uint8x16_t v4_2 = vld1q_u8(i2); const uint8_t *i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); if XNN_UNPREDICTABLE(bh < 4) { i3 = i2; } const uint8x16_t v4_3 = vld1q_u8(i3); const uint8_t *i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); if XNN_UNPREDICTABLE(bh <= 4) { i4 = i3; } const uint8x16_t v4_4 = vld1q_u8(i4); const uint8_t *i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); if XNN_UNPREDICTABLE(bh < 6) { i5 = i4; } const uint8x16_t v4_5 = vld1q_u8(i5); const uint8_t *i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); if XNN_UNPREDICTABLE(bh <= 6) { i6 = i5; } const uint8x16_t v4_6 = vld1q_u8(i6); const uint8_t *i7 = (const uint8_t*) ((uintptr_t) i6 + input_stride); if XNN_UNPREDICTABLE(bh < 8) { i7 = i6; } const uint8x16_t v4_7 = vld1q_u8(i7); const uint8_t *i8 = (const uint8_t*) ((uintptr_t) i7 + input_stride); if XNN_UNPREDICTABLE(bh <= 8) { i8 = i7; } const uint8x16_t v4_8 = vld1q_u8(i8); const uint8_t *i9 = (const uint8_t*) ((uintptr_t) i8 + input_stride); if XNN_UNPREDICTABLE(bh < 10) { i9 = i8; } const uint8x16_t v4_9 = vld1q_u8(i9); const uint8_t *i10 = (const uint8_t*) ((uintptr_t) i9 + input_stride); if XNN_UNPREDICTABLE(bh <= 10) { i10 = i9; } const uint8x16_t v4_10 = vld1q_u8(i10); const uint8_t *i11 = (const uint8_t*) ((uintptr_t) i10 + input_stride); if XNN_UNPREDICTABLE(bh < 12) { i11 = i10; } const uint8x16_t v4_11 = vld1q_u8(i11); const uint8_t *i12 = (const uint8_t*) ((uintptr_t) i11 + input_stride); if XNN_UNPREDICTABLE(bh <= 12) { i12 = i11; } const uint8x16_t v4_12 = vld1q_u8(i12); const uint8_t *i13 = (const uint8_t*) ((uintptr_t) i12 + input_stride); if XNN_UNPREDICTABLE(bh < 14) { i13 = i12; } const uint8x16_t v4_13 = vld1q_u8(i13); const uint8_t *i14 = (const uint8_t*) ((uintptr_t) i13 + input_stride); if XNN_UNPREDICTABLE(bh <= 14) { i14 = i13; } const uint8x16_t v4_14 = vld1q_u8(i14); const uint8x16_t v4_15 = vmovq_n_u8(0); const uint8x16x2_t v3_0 = vzipq_u8(v4_0, v4_8); const uint8x16x2_t v3_1 = vzipq_u8(v4_1, v4_9); const uint8x16x2_t v3_2 = vzipq_u8(v4_2, v4_10); const uint8x16x2_t v3_3 = vzipq_u8(v4_3, v4_11); const uint8x16x2_t v3_4 = vzipq_u8(v4_4, v4_12); const uint8x16x2_t v3_5 = vzipq_u8(v4_5, v4_13); const uint8x16x2_t v3_6 = vzipq_u8(v4_6, v4_14); const uint8x16x2_t v3_7 = vzipq_u8(v4_7, v4_15); const uint8x16x2_t v2_0 = vzipq_u8(v3_0.val[0], v3_4.val[0]); const uint8x16x2_t v2_1 = vzipq_u8(v3_0.val[1], v3_4.val[1]); const uint8x16x2_t v2_2 = vzipq_u8(v3_1.val[0], v3_5.val[0]); const uint8x16x2_t v2_3 = vzipq_u8(v3_1.val[1], v3_5.val[1]); const uint8x16x2_t v2_4 = vzipq_u8(v3_2.val[0], v3_6.val[0]); const uint8x16x2_t v2_5 = vzipq_u8(v3_2.val[1], v3_6.val[1]); const uint8x16x2_t v2_6 = vzipq_u8(v3_3.val[0], v3_7.val[0]); const uint8x16x2_t v2_7 = vzipq_u8(v3_3.val[1], v3_7.val[1]); const uint8x16x2_t v1_0 = vzipq_u8(v2_0.val[0], v2_4.val[0]); const uint8x16x2_t v1_1 = vzipq_u8(v2_0.val[1], v2_4.val[1]); const uint8x16x2_t v1_2 = vzipq_u8(v2_1.val[0], v2_5.val[0]); const uint8x16x2_t v1_3 = vzipq_u8(v2_1.val[1], v2_5.val[1]); const uint8x16x2_t v1_4 = vzipq_u8(v2_2.val[0], v2_6.val[0]); const uint8x16x2_t v1_5 = vzipq_u8(v2_2.val[1], v2_6.val[1]); const uint8x16x2_t v1_6 = vzipq_u8(v2_3.val[0], v2_7.val[0]); const uint8x16x2_t v1_7 = vzipq_u8(v2_3.val[1], v2_7.val[1]); const uint8x16x2_t v0_0 = vzipq_u8(v1_0.val[0], v1_4.val[0]); const uint8x16x2_t v0_1 = vzipq_u8(v1_0.val[1], v1_4.val[1]); const uint8x16x2_t v0_2 = vzipq_u8(v1_1.val[0], v1_5.val[0]); const uint8x16x2_t v0_3 = vzipq_u8(v1_1.val[1], v1_5.val[1]); const uint8x16x2_t v0_4 = vzipq_u8(v1_2.val[0], v1_6.val[0]); const uint8x16x2_t v0_5 = vzipq_u8(v1_2.val[1], v1_6.val[1]); const uint8x16x2_t v0_6 = vzipq_u8(v1_3.val[0], v1_7.val[0]); const uint8x16x2_t v0_7 = vzipq_u8(v1_3.val[1], v1_7.val[1]); uint8x8_t v0_low = vget_low_u8(v0_0.val[0]); uint8x8_t v1_low = vget_low_u8(v0_0.val[1]); uint8x8_t v2_low = vget_low_u8(v0_1.val[0]); uint8x8_t v3_low = vget_low_u8(v0_1.val[1]); uint8x8_t v4_low = vget_low_u8(v0_2.val[0]); uint8x8_t v5_low = vget_low_u8(v0_2.val[1]); uint8x8_t v6_low = vget_low_u8(v0_3.val[0]); uint8x8_t v7_low = vget_low_u8(v0_3.val[1]); uint8x8_t v8_low = vget_low_u8(v0_4.val[0]); uint8x8_t v9_low = vget_low_u8(v0_4.val[1]); uint8x8_t v10_low = vget_low_u8(v0_5.val[0]); uint8x8_t v11_low = vget_low_u8(v0_5.val[1]); uint8x8_t v12_low = vget_low_u8(v0_6.val[0]); uint8x8_t v13_low = vget_low_u8(v0_6.val[1]); uint8x8_t v14_low = vget_low_u8(v0_7.val[0]); uint8x8_t v15_low = vget_low_u8(v0_7.val[1]); if (bh & 8) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 15: vst1_u8(oN, v15_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 14: vst1_u8(oN, v14_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 13: vst1_u8(oN, v13_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 12: vst1_u8(oN, v12_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 11: vst1_u8(oN, v11_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 10: vst1_u8(oN, v10_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 9: vst1_u8(oN, v9_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 8: vst1_u8(oN, v8_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 7: vst1_u8(oN, v7_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_u8(oN, v6_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_u8(oN, v5_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_u8(oN, v4_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_u8(oN, v3_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_u8(oN, v2_low); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_u8(oN, v1_low); case 0: vst1_u8(o, v0_low); o += 8; break; default: XNN_UNREACHABLE; } v0_low = vget_high_u8(v0_0.val[0]); v1_low = vget_high_u8(v0_0.val[1]); v2_low = vget_high_u8(v0_1.val[0]); v3_low = vget_high_u8(v0_1.val[1]); v4_low = vget_high_u8(v0_2.val[0]); v5_low = vget_high_u8(v0_2.val[1]); v6_low = vget_high_u8(v0_3.val[0]); v7_low = vget_high_u8(v0_3.val[1]); v8_low = vget_high_u8(v0_4.val[0]); v9_low = vget_high_u8(v0_4.val[1]); v10_low = vget_high_u8(v0_5.val[0]); v11_low = vget_high_u8(v0_5.val[1]); v12_low = vget_high_u8(v0_6.val[0]); v13_low = vget_high_u8(v0_6.val[1]); v14_low = vget_high_u8(v0_7.val[0]); v15_low = vget_high_u8(v0_7.val[1]); } if (bh & 4) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 15: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v15_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 14: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v14_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 13: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v13_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 12: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v12_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 11: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v11_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 10: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v10_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 9: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v9_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 8: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v8_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 7: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v7_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v1_low), 0); case 0: vst1_lane_u32((void*) o, vreinterpret_u32_u8(v0_low), 0); o += 4; break; default: XNN_UNREACHABLE; } v0_low = vext_u8(v0_low, v0_low, 4); v1_low = vext_u8(v1_low, v1_low, 4); v2_low = vext_u8(v2_low, v2_low, 4); v3_low = vext_u8(v3_low, v3_low, 4); v4_low = vext_u8(v4_low, v4_low, 4); v5_low = vext_u8(v5_low, v5_low, 4); v6_low = vext_u8(v6_low, v6_low, 4); v7_low = vext_u8(v7_low, v7_low, 4); v8_low = vext_u8(v8_low, v8_low, 4); v9_low = vext_u8(v9_low, v9_low, 4); v10_low = vext_u8(v10_low, v10_low, 4); v11_low = vext_u8(v11_low, v11_low, 4); v12_low = vext_u8(v12_low, v12_low, 4); v13_low = vext_u8(v13_low, v13_low, 4); v14_low = vext_u8(v14_low, v14_low, 4); v15_low = vext_u8(v15_low, v15_low, 4); } if (bh & 2) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 15: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v15_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 14: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v14_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 13: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v13_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 12: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v12_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 11: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v11_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 10: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v10_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 9: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v9_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 8: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v8_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 7: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v7_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v1_low), 0); case 0: vst1_lane_u16((void*) o, vreinterpret_u16_u8(v0_low), 0); o += 2; break; default: XNN_UNREACHABLE; } v0_low = vext_u8(v0_low, v0_low, 2); v1_low = vext_u8(v1_low, v1_low, 2); v2_low = vext_u8(v2_low, v2_low, 2); v3_low = vext_u8(v3_low, v3_low, 2); v4_low = vext_u8(v4_low, v4_low, 2); v5_low = vext_u8(v5_low, v5_low, 2); v6_low = vext_u8(v6_low, v6_low, 2); v7_low = vext_u8(v7_low, v7_low, 2); v8_low = vext_u8(v8_low, v8_low, 2); v9_low = vext_u8(v9_low, v9_low, 2); v10_low = vext_u8(v10_low, v10_low, 2); v11_low = vext_u8(v11_low, v11_low, 2); v12_low = vext_u8(v12_low, v12_low, 2); v13_low = vext_u8(v13_low, v13_low, 2); v14_low = vext_u8(v14_low, v14_low, 2); v15_low = vext_u8(v15_low, v15_low, 2); } if (bh & 1) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 15: vst1_lane_u8(oN, v15_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 14: vst1_lane_u8(oN, v14_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 13: vst1_lane_u8(oN, v13_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 12: vst1_lane_u8(oN, v12_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 11: vst1_lane_u8(oN, v11_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 10: vst1_lane_u8(oN, v10_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 9: vst1_lane_u8(oN, v9_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 8: vst1_lane_u8(oN, v8_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 7: vst1_lane_u8(oN, v7_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_lane_u8(oN, v6_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_lane_u8(oN, v5_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_lane_u8(oN, v4_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_lane_u8(oN, v3_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_lane_u8(oN, v2_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_lane_u8(oN, v1_low, 0); case 0: vst1_lane_u8(o, v0_low, 0); break; default: XNN_UNREACHABLE; } } } i0 = (const uint8_t*) ((uintptr_t) i0 + input_reset); o = (uint8_t*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
23,468
48.098326
127
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-1x2-scalar-int.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__1x2_scalar_int( const uint8_t *input, uint8_t * output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params* params) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(int8_t)); assert(input_stride >= block_width * sizeof(int8_t)); const size_t tile_height = 1; const size_t tile_width = 2; const size_t tile_wbytes = tile_width * sizeof(int8_t); const size_t input_reset = tile_wbytes - block_height * input_stride; const size_t output_reset = tile_width * output_stride - block_height * sizeof(int8_t); const size_t input_offset = tile_height * input_stride; const int8_t* i0 = (const int8_t*) input; int8_t* o0 = (int8_t*) output; int8_t* o1 = (int8_t*) ((uintptr_t) o0 + output_stride); do { if XNN_UNPREDICTABLE(block_width < 2) { o1 = o0; } size_t bh = block_height; for (; bh >= 1; bh -= 1) { *o1++ = i0[1]; *o0++ = i0[0]; i0 = (const int8_t*) ((uintptr_t) i0 + input_offset); } i0 = (const int8_t*) ((uintptr_t) i0 + input_reset); o0 = (int8_t*) ((uintptr_t) o0 + output_reset); o1 = (int8_t*) ((uintptr_t) o1 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
1,726
28.775862
89
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-1x4-scalar-int.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__1x4_scalar_int( const uint8_t *input, uint8_t * output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params* params) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(int8_t)); assert(input_stride >= block_width * sizeof(int8_t)); const size_t tile_height = 1; const size_t tile_width = 4; const size_t tile_wbytes = tile_width * sizeof(int8_t); const size_t input_reset = tile_wbytes - block_height * input_stride; const size_t output_reset = tile_width * output_stride - block_height * sizeof(int8_t); const size_t input_offset = tile_height * input_stride; const int8_t* i0 = (const int8_t*) input; int8_t* o0 = (int8_t*) output; int8_t* o1 = (int8_t*) ((uintptr_t) o0 + output_stride); int8_t* o2 = (int8_t*) ((uintptr_t) o1 + output_stride); int8_t* o3 = (int8_t*) ((uintptr_t) o2 + output_stride); do { if XNN_UNPREDICTABLE(block_width < 2) { o1 = o0; } if XNN_UNPREDICTABLE(block_width <= 2) { o2 = o0; } if XNN_UNPREDICTABLE(block_width < 4) { o3 = o0; } size_t bh = block_height; for (; bh >= 1; bh -= 1) { *o3++ = i0[3]; *o2++ = i0[2]; *o1++ = i0[1]; *o0++ = i0[0]; i0 = (const int8_t*) ((uintptr_t) i0 + input_offset); } i0 = (const int8_t*) ((uintptr_t) i0 + input_reset); o0 = (int8_t*) ((uintptr_t) o0 + output_reset); o1 = (int8_t*) ((uintptr_t) o1 + output_reset); o2 = (int8_t*) ((uintptr_t) o2 + output_reset); o3 = (int8_t*) ((uintptr_t) o3 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
2,121
29.314286
89
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-2x1-scalar-int.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__2x1_scalar_int( const uint8_t *input, uint8_t * output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params* params) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(int8_t)); assert(input_stride >= block_width * sizeof(int8_t)); const size_t tile_height = 2; const size_t tile_width = 1; const size_t tile_wbytes = tile_width * sizeof(int8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(int8_t); const size_t input_offset = tile_height * input_stride; const int8_t* i0 = (const int8_t*) input; const int8_t* i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); int8_t* o0 = (int8_t*) output; do { size_t bh = block_height; for (; bh >= 2; bh -= 2) { *o0++ = i0[0]; *o0++ = i1[0]; i0 = (const int8_t*) ((uintptr_t) i0 + input_offset); i1 = (const int8_t*) ((uintptr_t) i1 + input_offset); } if (bh & 1) { o0[0] = i0[0]; } i0 = (const int8_t*) ((uintptr_t) i0 + input_reset); i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); o0 = (int8_t*) ((uintptr_t) o0 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
1,831
30.050847
108
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-2x2-scalar-int.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__2x2_scalar_int( const uint8_t *input, uint8_t * output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params* params) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(int8_t)); assert(input_stride >= block_width * sizeof(int8_t)); const size_t tile_height = 2; const size_t tile_width = 2; const size_t tile_wbytes = tile_width * sizeof(int8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(int8_t); const size_t input_offset = tile_height * input_stride; const int8_t* i0 = (const int8_t*) input; const int8_t* i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); int8_t* o0 = (int8_t*) output; int8_t* o1 = (int8_t*) ((uintptr_t) o0 + output_stride); do { if XNN_UNPREDICTABLE(block_width < 2) { o1 = o0; } size_t bh = block_height; for (; bh >= 2; bh -= 2) { *o1++ = i0[1]; *o1++ = i1[1]; *o0++ = i0[0]; *o0++ = i1[0]; i0 = (const int8_t*) ((uintptr_t) i0 + input_offset); i1 = (const int8_t*) ((uintptr_t) i1 + input_offset); } if (bh & 1) { o1[0] = i0[1]; o0[0] = i0[0]; } i0 = (const int8_t*) ((uintptr_t) i0 + input_reset); i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); o0 = (int8_t*) ((uintptr_t) o0 + output_reset); o1 = (int8_t*) ((uintptr_t) o1 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
2,070
29.910448
108
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-2x4-scalar-int.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__2x4_scalar_int( const uint8_t *input, uint8_t * output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params* params) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(int8_t)); assert(input_stride >= block_width * sizeof(int8_t)); const size_t tile_height = 2; const size_t tile_width = 4; const size_t tile_wbytes = tile_width * sizeof(int8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(int8_t); const size_t input_offset = tile_height * input_stride; const int8_t* i0 = (const int8_t*) input; const int8_t* i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); int8_t* o0 = (int8_t*) output; int8_t* o1 = (int8_t*) ((uintptr_t) o0 + output_stride); int8_t* o2 = (int8_t*) ((uintptr_t) o1 + output_stride); int8_t* o3 = (int8_t*) ((uintptr_t) o2 + output_stride); do { if XNN_UNPREDICTABLE(block_width < 2) { o1 = o0; } if XNN_UNPREDICTABLE(block_width <= 2) { o2 = o0; } if XNN_UNPREDICTABLE(block_width < 4) { o3 = o0; } size_t bh = block_height; for (; bh >= 2; bh -= 2) { *o3++ = i0[3]; *o3++ = i1[3]; *o2++ = i0[2]; *o2++ = i1[2]; *o1++ = i0[1]; *o1++ = i1[1]; *o0++ = i0[0]; *o0++ = i1[0]; i0 = (const int8_t*) ((uintptr_t) i0 + input_offset); i1 = (const int8_t*) ((uintptr_t) i1 + input_offset); } if (bh & 1) { o3[0] = i0[3]; o2[0] = i0[2]; o1[0] = i0[1]; o0[0] = i0[0]; } i0 = (const int8_t*) ((uintptr_t) i0 + input_reset); i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); o0 = (int8_t*) ((uintptr_t) o0 + output_reset); o1 = (int8_t*) ((uintptr_t) o1 + output_reset); o2 = (int8_t*) ((uintptr_t) o2 + output_reset); o3 = (int8_t*) ((uintptr_t) o3 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
2,549
29.722892
108
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-4x1-scalar-int.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__4x1_scalar_int( const uint8_t *input, uint8_t * output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params* params) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(int8_t)); assert(input_stride >= block_width * sizeof(int8_t)); const size_t tile_height = 4; const size_t tile_width = 1; const size_t tile_wbytes = tile_width * sizeof(int8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(int8_t); const size_t input_offset = tile_height * input_stride; const int8_t* i0 = (const int8_t*) input; const int8_t* i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); const int8_t* i2 = (const int8_t*) ((uintptr_t) i1 + input_stride); const int8_t* i3 = (const int8_t*) ((uintptr_t) i2 + input_stride); int8_t* o0 = (int8_t*) output; do { size_t bh = block_height; for (; bh >= 4; bh -= 4) { *o0++ = i0[0]; *o0++ = i1[0]; *o0++ = i2[0]; *o0++ = i3[0]; i0 = (const int8_t*) ((uintptr_t) i0 + input_offset); i1 = (const int8_t*) ((uintptr_t) i1 + input_offset); i2 = (const int8_t*) ((uintptr_t) i2 + input_offset); i3 = (const int8_t*) ((uintptr_t) i3 + input_offset); } const int8_t* i = i0; if (bh & 2) { o0[0] = i0[0]; o0[1] = i1[0]; o0 += 2; i = i2; } if (bh & 1) { o0[0] = i[0]; } i0 = (const int8_t*) ((uintptr_t) i0 + input_reset); i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); i2 = (const int8_t*) ((uintptr_t) i1 + input_stride); i3 = (const int8_t*) ((uintptr_t) i2 + input_stride); o0 = (int8_t*) ((uintptr_t) o0 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
2,369
31.027027
108
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-4x2-scalar-int.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__4x2_scalar_int( const uint8_t *input, uint8_t * output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params* params) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(int8_t)); assert(input_stride >= block_width * sizeof(int8_t)); const size_t tile_height = 4; const size_t tile_width = 2; const size_t tile_wbytes = tile_width * sizeof(int8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(int8_t); const size_t input_offset = tile_height * input_stride; const int8_t* i0 = (const int8_t*) input; const int8_t* i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); const int8_t* i2 = (const int8_t*) ((uintptr_t) i1 + input_stride); const int8_t* i3 = (const int8_t*) ((uintptr_t) i2 + input_stride); int8_t* o0 = (int8_t*) output; int8_t* o1 = (int8_t*) ((uintptr_t) o0 + output_stride); do { if XNN_UNPREDICTABLE(block_width < 2) { o1 = o0; } size_t bh = block_height; for (; bh >= 4; bh -= 4) { *o1++ = i0[1]; *o1++ = i1[1]; *o1++ = i2[1]; *o1++ = i3[1]; *o0++ = i0[0]; *o0++ = i1[0]; *o0++ = i2[0]; *o0++ = i3[0]; i0 = (const int8_t*) ((uintptr_t) i0 + input_offset); i1 = (const int8_t*) ((uintptr_t) i1 + input_offset); i2 = (const int8_t*) ((uintptr_t) i2 + input_offset); i3 = (const int8_t*) ((uintptr_t) i3 + input_offset); } const int8_t* i = i0; if (bh & 2) { o1[0] = i0[1]; o1[1] = i1[1]; o1 += 2; o0[0] = i0[0]; o0[1] = i1[0]; o0 += 2; i = i2; } if (bh & 1) { o1[0] = i[1]; o0[0] = i[0]; } i0 = (const int8_t*) ((uintptr_t) i0 + input_reset); i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); i2 = (const int8_t*) ((uintptr_t) i1 + input_stride); i3 = (const int8_t*) ((uintptr_t) i2 + input_stride); o0 = (int8_t*) ((uintptr_t) o0 + output_reset); o1 = (int8_t*) ((uintptr_t) o1 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
2,706
30.114943
108
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-4x4-scalar-int.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/scalar.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__4x4_scalar_int( const uint8_t *input, uint8_t * output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params* params) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(int8_t)); assert(input_stride >= block_width * sizeof(int8_t)); const size_t tile_height = 4; const size_t tile_width = 4; const size_t tile_wbytes = tile_width * sizeof(int8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(int8_t); const size_t input_offset = tile_height * input_stride; const int8_t* i0 = (const int8_t*) input; const int8_t* i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); const int8_t* i2 = (const int8_t*) ((uintptr_t) i1 + input_stride); const int8_t* i3 = (const int8_t*) ((uintptr_t) i2 + input_stride); int8_t* o0 = (int8_t*) output; int8_t* o1 = (int8_t*) ((uintptr_t) o0 + output_stride); int8_t* o2 = (int8_t*) ((uintptr_t) o1 + output_stride); int8_t* o3 = (int8_t*) ((uintptr_t) o2 + output_stride); do { if XNN_UNPREDICTABLE(block_width < 2) { o1 = o0; } if XNN_UNPREDICTABLE(block_width <= 2) { o2 = o0; } if XNN_UNPREDICTABLE(block_width < 4) { o3 = o0; } size_t bh = block_height; for (; bh >= 4; bh -= 4) { *o3++ = i0[3]; *o3++ = i1[3]; *o3++ = i2[3]; *o3++ = i3[3]; *o2++ = i0[2]; *o2++ = i1[2]; *o2++ = i2[2]; *o2++ = i3[2]; *o1++ = i0[1]; *o1++ = i1[1]; *o1++ = i2[1]; *o1++ = i3[1]; *o0++ = i0[0]; *o0++ = i1[0]; *o0++ = i2[0]; *o0++ = i3[0]; i0 = (const int8_t*) ((uintptr_t) i0 + input_offset); i1 = (const int8_t*) ((uintptr_t) i1 + input_offset); i2 = (const int8_t*) ((uintptr_t) i2 + input_offset); i3 = (const int8_t*) ((uintptr_t) i3 + input_offset); } const int8_t* i = i0; if (bh & 2) { o3[0] = i0[3]; o3[1] = i1[3]; o3 += 2; o2[0] = i0[2]; o2[1] = i1[2]; o2 += 2; o1[0] = i0[1]; o1[1] = i1[1]; o1 += 2; o0[0] = i0[0]; o0[1] = i1[0]; o0 += 2; i = i2; } if (bh & 1) { o3[0] = i[3]; o2[0] = i[2]; o1[0] = i[1]; o0[0] = i[0]; } i0 = (const int8_t*) ((uintptr_t) i0 + input_reset); i1 = (const int8_t*) ((uintptr_t) i0 + input_stride); i2 = (const int8_t*) ((uintptr_t) i1 + input_stride); i3 = (const int8_t*) ((uintptr_t) i2 + input_stride); o0 = (int8_t*) ((uintptr_t) o0 + output_reset); o1 = (int8_t*) ((uintptr_t) o1 + output_reset); o2 = (int8_t*) ((uintptr_t) o2 + output_reset); o3 = (int8_t*) ((uintptr_t) o3 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
3,381
28.929204
108
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-8x8-multi-dec-zip-neon.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/neon-zip.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__8x8_multi_dec_zip_neon( const uint8_t* input, uint8_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(uint8_t)); assert(input_stride >= block_width * sizeof(uint8_t)); const size_t tile_height = 8; const size_t tile_width = 8; const size_t tile_hbytes = tile_height * sizeof(uint8_t); const size_t tile_wbytes = tile_width * sizeof(uint8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t input_offset = tile_height * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(uint8_t) - tile_hbytes; const uint8_t* i0 = input; const uint8_t* i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); const uint8_t* i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); const uint8_t* i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); const uint8_t* i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); const uint8_t* i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); const uint8_t* i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); const uint8_t* i7 = (const uint8_t*) ((uintptr_t) i6 + input_stride); uint8_t* o = (uint8_t*) ((uintptr_t) output - tile_hbytes); const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 7); const size_t oN_stride = rem * output_stride; const size_t oN_offset = oN_stride + tile_hbytes; size_t bh = block_height; for (; bh >= 8; bh -= 8) { const uint8x8_t v3_0 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_offset); const uint8x8_t v3_1 = vld1_u8(i1); i1 = (uint8_t*) ((uintptr_t) i1 + input_offset); const uint8x8_t v3_2 = vld1_u8(i2); i2 = (uint8_t*) ((uintptr_t) i2 + input_offset); const uint8x8_t v3_3 = vld1_u8(i3); i3 = (uint8_t*) ((uintptr_t) i3 + input_offset); const uint8x8_t v3_4 = vld1_u8(i4); i4 = (uint8_t*) ((uintptr_t) i4 + input_offset); const uint8x8_t v3_5 = vld1_u8(i5); i5 = (uint8_t*) ((uintptr_t) i5 + input_offset); const uint8x8_t v3_6 = vld1_u8(i6); i6 = (uint8_t*) ((uintptr_t) i6 + input_offset); const uint8x8_t v3_7 = vld1_u8(i7); i7 = (uint8_t*) ((uintptr_t) i7 + input_offset); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); o = (uint8_t*) ((uintptr_t) o + oN_offset); vst1_u8(o, v0_3.val[1]); if XNN_UNPREDICTABLE(block_width > 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_3.val[0]); if XNN_UNPREDICTABLE(block_width >= 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_2.val[1]); if XNN_UNPREDICTABLE(block_width > 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_2.val[0]); if XNN_UNPREDICTABLE(block_width >= 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_1.val[1]); if XNN_UNPREDICTABLE(block_width > 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_1.val[0]); if XNN_UNPREDICTABLE(block_width >= 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_0.val[1]); if XNN_UNPREDICTABLE(block_width > 1) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_0.val[0]); } o = (uint8_t*) ((uintptr_t) o + tile_hbytes); if (bh != 0) { const uint8x8_t v3_0 = vld1_u8(i0); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const uint8x8_t v3_1 = vld1_u8(i1); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i0; } const uint8x8_t v3_2 = vld1_u8(i2); if XNN_UNPREDICTABLE(bh < 4) { i3 = i0; } const uint8x8_t v3_3 = vld1_u8(i3); if XNN_UNPREDICTABLE(bh <= 4) { i4 = i0; } const uint8x8_t v3_4 = vld1_u8(i4); if XNN_UNPREDICTABLE(bh < 6) { i5 = i0; } const uint8x8_t v3_5 = vld1_u8(i5); if XNN_UNPREDICTABLE(bh <= 6) { i6 = i0; } const uint8x8_t v3_6 = vld1_u8(i6); const uint8x8_t v3_7 = vmov_n_u8(0); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); uint8x8_t v0_low = v0_0.val[0]; uint8x8_t v1_low = v0_0.val[1]; uint8x8_t v2_low = v0_1.val[0]; uint8x8_t v3_low = v0_1.val[1]; uint8x8_t v4_low = v0_2.val[0]; uint8x8_t v5_low = v0_2.val[1]; uint8x8_t v6_low = v0_3.val[0]; uint8x8_t v7_low = v0_3.val[1]; if (bh & 4) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u32((void*) o, vreinterpret_u32_u8(v7_low), 0); if XNN_UNPREDICTABLE(block_width > 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v6_low), 0); if XNN_UNPREDICTABLE(block_width >= 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v5_low), 0); if XNN_UNPREDICTABLE(block_width > 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v4_low), 0); if XNN_UNPREDICTABLE(block_width >= 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v3_low), 0); if XNN_UNPREDICTABLE(block_width > 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v2_low), 0); if XNN_UNPREDICTABLE(block_width >= 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v1_low), 0); if XNN_UNPREDICTABLE(block_width > 1) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v0_low), 0); o += 4; v0_low = vext_u8(v0_low, v0_low, 4); v1_low = vext_u8(v1_low, v1_low, 4); v2_low = vext_u8(v2_low, v2_low, 4); v3_low = vext_u8(v3_low, v3_low, 4); v4_low = vext_u8(v4_low, v4_low, 4); v5_low = vext_u8(v5_low, v5_low, 4); v6_low = vext_u8(v6_low, v6_low, 4); v7_low = vext_u8(v7_low, v7_low, 4); } if (bh & 2) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u16((void*) o, vreinterpret_u16_u8(v7_low), 0); if XNN_UNPREDICTABLE(block_width > 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v6_low), 0); if XNN_UNPREDICTABLE(block_width >= 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v5_low), 0); if XNN_UNPREDICTABLE(block_width > 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v4_low), 0); if XNN_UNPREDICTABLE(block_width >= 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v3_low), 0); if XNN_UNPREDICTABLE(block_width > 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v2_low), 0); if XNN_UNPREDICTABLE(block_width >= 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v1_low), 0); if XNN_UNPREDICTABLE(block_width > 1) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v0_low), 0); o += 2; v0_low = vext_u8(v0_low, v0_low, 2); v1_low = vext_u8(v1_low, v1_low, 2); v2_low = vext_u8(v2_low, v2_low, 2); v3_low = vext_u8(v3_low, v3_low, 2); v4_low = vext_u8(v4_low, v4_low, 2); v5_low = vext_u8(v5_low, v5_low, 2); v6_low = vext_u8(v6_low, v6_low, 2); v7_low = vext_u8(v7_low, v7_low, 2); } if (bh & 1) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u8(o, v7_low, 0); if XNN_UNPREDICTABLE(block_width > 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v6_low, 0); if XNN_UNPREDICTABLE(block_width >= 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v5_low, 0); if XNN_UNPREDICTABLE(block_width > 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v4_low, 0); if XNN_UNPREDICTABLE(block_width >= 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v3_low, 0); if XNN_UNPREDICTABLE(block_width > 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v2_low, 0); if XNN_UNPREDICTABLE(block_width >= 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v1_low, 0); if XNN_UNPREDICTABLE(block_width > 1) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v0_low, 0); } } i0 = (const uint8_t*) ((uintptr_t) i0 + input_reset); i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); i7 = (const uint8_t*) ((uintptr_t) i6 + input_stride); o = (uint8_t*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
12,072
40.920139
123
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-8x8-multi-mov-zip-neon.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/neon-zip.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__8x8_multi_mov_zip_neon( const uint8_t* input, uint8_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(uint8_t)); assert(input_stride >= block_width * sizeof(uint8_t)); const size_t tile_height = 8; const size_t tile_width = 8; const size_t tile_hbytes = tile_height * sizeof(uint8_t); const size_t tile_wbytes = tile_width * sizeof(uint8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t input_offset = tile_height * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(uint8_t) - tile_hbytes; const uint8_t* i0 = input; const uint8_t* i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); const uint8_t* i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); const uint8_t* i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); const uint8_t* i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); const uint8_t* i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); const uint8_t* i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); const uint8_t* i7 = (const uint8_t*) ((uintptr_t) i6 + input_stride); uint8_t* o = (uint8_t*) ((uintptr_t) output - tile_hbytes); const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 7); const size_t oN_stride = rem * output_stride; const size_t oN_offset = oN_stride + tile_hbytes; size_t bh = block_height; for (; bh >= 8; bh -= 8) { const uint8x8_t v3_0 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_offset); const uint8x8_t v3_1 = vld1_u8(i1); i1 = (uint8_t*) ((uintptr_t) i1 + input_offset); const uint8x8_t v3_2 = vld1_u8(i2); i2 = (uint8_t*) ((uintptr_t) i2 + input_offset); const uint8x8_t v3_3 = vld1_u8(i3); i3 = (uint8_t*) ((uintptr_t) i3 + input_offset); const uint8x8_t v3_4 = vld1_u8(i4); i4 = (uint8_t*) ((uintptr_t) i4 + input_offset); const uint8x8_t v3_5 = vld1_u8(i5); i5 = (uint8_t*) ((uintptr_t) i5 + input_offset); const uint8x8_t v3_6 = vld1_u8(i6); i6 = (uint8_t*) ((uintptr_t) i6 + input_offset); const uint8x8_t v3_7 = vld1_u8(i7); i7 = (uint8_t*) ((uintptr_t) i7 + input_offset); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); o = (uint8_t*) ((uintptr_t) o + oN_offset); vst1_u8(o, v0_3.val[1]); uint8_t *oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 7) { o = oN; } vst1_u8(o, v0_3.val[0]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 7) { o = oN; } vst1_u8(o, v0_2.val[1]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 5) { o = oN; } vst1_u8(o, v0_2.val[0]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 5) { o = oN; } vst1_u8(o, v0_1.val[1]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } vst1_u8(o, v0_1.val[0]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } vst1_u8(o, v0_0.val[1]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } vst1_u8(o, v0_0.val[0]); } o = (uint8_t*) ((uintptr_t) o + tile_hbytes); if (bh != 0) { const uint8x8_t v3_0 = vld1_u8(i0); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const uint8x8_t v3_1 = vld1_u8(i1); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i0; } const uint8x8_t v3_2 = vld1_u8(i2); if XNN_UNPREDICTABLE(bh < 4) { i3 = i0; } const uint8x8_t v3_3 = vld1_u8(i3); if XNN_UNPREDICTABLE(bh <= 4) { i4 = i0; } const uint8x8_t v3_4 = vld1_u8(i4); if XNN_UNPREDICTABLE(bh < 6) { i5 = i0; } const uint8x8_t v3_5 = vld1_u8(i5); if XNN_UNPREDICTABLE(bh <= 6) { i6 = i0; } const uint8x8_t v3_6 = vld1_u8(i6); const uint8x8_t v3_7 = vmov_n_u8(0); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); uint8x8_t v0_low = v0_0.val[0]; uint8x8_t v1_low = v0_0.val[1]; uint8x8_t v2_low = v0_1.val[0]; uint8x8_t v3_low = v0_1.val[1]; uint8x8_t v4_low = v0_2.val[0]; uint8x8_t v5_low = v0_2.val[1]; uint8x8_t v6_low = v0_3.val[0]; uint8x8_t v7_low = v0_3.val[1]; if (bh & 4) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u32((void*) o, vreinterpret_u32_u8(v7_low), 0); uint8_t *oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 7) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 7) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 5) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 5) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v1_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v0_low), 0); o += 4; v0_low = vext_u8(v0_low, v0_low, 4); v1_low = vext_u8(v1_low, v1_low, 4); v2_low = vext_u8(v2_low, v2_low, 4); v3_low = vext_u8(v3_low, v3_low, 4); v4_low = vext_u8(v4_low, v4_low, 4); v5_low = vext_u8(v5_low, v5_low, 4); v6_low = vext_u8(v6_low, v6_low, 4); v7_low = vext_u8(v7_low, v7_low, 4); } if (bh & 2) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u16((void*) o, vreinterpret_u16_u8(v7_low), 0); uint8_t *oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 7) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 7) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 5) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 5) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v1_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v0_low), 0); o += 2; v0_low = vext_u8(v0_low, v0_low, 2); v1_low = vext_u8(v1_low, v1_low, 2); v2_low = vext_u8(v2_low, v2_low, 2); v3_low = vext_u8(v3_low, v3_low, 2); v4_low = vext_u8(v4_low, v4_low, 2); v5_low = vext_u8(v5_low, v5_low, 2); v6_low = vext_u8(v6_low, v6_low, 2); v7_low = vext_u8(v7_low, v7_low, 2); } if (bh & 1) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u8(o, v7_low, 0); uint8_t *oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 7) { o = oN; } vst1_lane_u8(o, v6_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 7) { o = oN; } vst1_lane_u8(o, v5_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 5) { o = oN; } vst1_lane_u8(o, v4_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 5) { o = oN; } vst1_lane_u8(o, v3_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } vst1_lane_u8(o, v2_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } vst1_lane_u8(o, v1_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } vst1_lane_u8(o, v0_low, 0); } } i0 = (const uint8_t*) ((uintptr_t) i0 + input_reset); i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); i7 = (const uint8_t*) ((uintptr_t) i6 + input_stride); o = (uint8_t*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
12,570
38.781646
123
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-8x8-multi-switch-zip-neon.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/neon-zip.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__8x8_multi_switch_zip_neon( const uint8_t* input, uint8_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(uint8_t)); assert(input_stride >= block_width * sizeof(uint8_t)); const size_t tile_height = 8; const size_t tile_width = 8; const size_t tile_hbytes = tile_height * sizeof(uint8_t); const size_t tile_wbytes = tile_width * sizeof(uint8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t input_offset = tile_height * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(uint8_t); const uint8_t* i0 = input; const uint8_t* i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); const uint8_t* i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); const uint8_t* i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); const uint8_t* i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); const uint8_t* i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); const uint8_t* i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); const uint8_t* i7 = (const uint8_t*) ((uintptr_t) i6 + input_stride); uint8_t* o = (uint8_t*) output; const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 7); const size_t oN_stride = rem * output_stride; size_t bh = block_height; for (; bh >= 8; bh -= 8) { const uint8x8_t v3_0 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_offset); const uint8x8_t v3_1 = vld1_u8(i1); i1 = (uint8_t*) ((uintptr_t) i1 + input_offset); const uint8x8_t v3_2 = vld1_u8(i2); i2 = (uint8_t*) ((uintptr_t) i2 + input_offset); const uint8x8_t v3_3 = vld1_u8(i3); i3 = (uint8_t*) ((uintptr_t) i3 + input_offset); const uint8x8_t v3_4 = vld1_u8(i4); i4 = (uint8_t*) ((uintptr_t) i4 + input_offset); const uint8x8_t v3_5 = vld1_u8(i5); i5 = (uint8_t*) ((uintptr_t) i5 + input_offset); const uint8x8_t v3_6 = vld1_u8(i6); i6 = (uint8_t*) ((uintptr_t) i6 + input_offset); const uint8x8_t v3_7 = vld1_u8(i7); i7 = (uint8_t*) ((uintptr_t) i7 + input_offset); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); uint8_t *oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 7: vst1_u8(oN, v0_3.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_u8(oN, v0_3.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_u8(oN, v0_2.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_u8(oN, v0_2.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_u8(oN, v0_1.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_u8(oN, v0_1.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_u8(oN, v0_0.val[1]); case 0: vst1_u8(o, v0_0.val[0]); o = (uint8_t*) ((uintptr_t) o + tile_hbytes); break; default: XNN_UNREACHABLE; } } if (bh != 0) { const uint8x8_t v3_0 = vld1_u8(i0); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const uint8x8_t v3_1 = vld1_u8(i1); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i0; } const uint8x8_t v3_2 = vld1_u8(i2); if XNN_UNPREDICTABLE(bh < 4) { i3 = i0; } const uint8x8_t v3_3 = vld1_u8(i3); if XNN_UNPREDICTABLE(bh <= 4) { i4 = i0; } const uint8x8_t v3_4 = vld1_u8(i4); if XNN_UNPREDICTABLE(bh < 6) { i5 = i0; } const uint8x8_t v3_5 = vld1_u8(i5); if XNN_UNPREDICTABLE(bh <= 6) { i6 = i0; } const uint8x8_t v3_6 = vld1_u8(i6); const uint8x8_t v3_7 = vmov_n_u8(0); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); uint8x8_t v0_low = v0_0.val[0]; uint8x8_t v1_low = v0_0.val[1]; uint8x8_t v2_low = v0_1.val[0]; uint8x8_t v3_low = v0_1.val[1]; uint8x8_t v4_low = v0_2.val[0]; uint8x8_t v5_low = v0_2.val[1]; uint8x8_t v6_low = v0_3.val[0]; uint8x8_t v7_low = v0_3.val[1]; if (bh & 4) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 7: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v7_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v1_low), 0); case 0: vst1_lane_u32((void*) o, vreinterpret_u32_u8(v0_low), 0); o += 4; break; default: XNN_UNREACHABLE; } v0_low = vext_u8(v0_low, v0_low, 4); v1_low = vext_u8(v1_low, v1_low, 4); v2_low = vext_u8(v2_low, v2_low, 4); v3_low = vext_u8(v3_low, v3_low, 4); v4_low = vext_u8(v4_low, v4_low, 4); v5_low = vext_u8(v5_low, v5_low, 4); v6_low = vext_u8(v6_low, v6_low, 4); v7_low = vext_u8(v7_low, v7_low, 4); } if (bh & 2) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 7: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v7_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v1_low), 0); case 0: vst1_lane_u16((void*) o, vreinterpret_u16_u8(v0_low), 0); o += 2; break; default: XNN_UNREACHABLE; } v0_low = vext_u8(v0_low, v0_low, 2); v1_low = vext_u8(v1_low, v1_low, 2); v2_low = vext_u8(v2_low, v2_low, 2); v3_low = vext_u8(v3_low, v3_low, 2); v4_low = vext_u8(v4_low, v4_low, 2); v5_low = vext_u8(v5_low, v5_low, 2); v6_low = vext_u8(v6_low, v6_low, 2); v7_low = vext_u8(v7_low, v7_low, 2); } if (bh & 1) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 7: vst1_lane_u8(oN, v7_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_lane_u8(oN, v6_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_lane_u8(oN, v5_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_lane_u8(oN, v4_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_lane_u8(oN, v3_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_lane_u8(oN, v2_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_lane_u8(oN, v1_low, 0); case 0: vst1_lane_u8(o, v0_low, 0); break; default: XNN_UNREACHABLE; } } } i0 = (const uint8_t*) ((uintptr_t) i0 + input_reset); i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); i7 = (const uint8_t*) ((uintptr_t) i6 + input_stride); o = (uint8_t*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
11,079
42.622047
126
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-8x8-reuse-dec-zip-neon.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/neon-zip.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__8x8_reuse_dec_zip_neon( const uint8_t* input, uint8_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(uint8_t)); assert(input_stride >= block_width * sizeof(uint8_t)); const size_t tile_height = 8; const size_t tile_width = 8; const size_t tile_hbytes = tile_height * sizeof(uint8_t); const size_t tile_wbytes = tile_width * sizeof(uint8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(uint8_t) - tile_hbytes; const uint8_t* i0 = input; uint8_t* o = (uint8_t*) ((uintptr_t) output - tile_hbytes); const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 7); const size_t oN_stride = rem * output_stride; const size_t oN_offset = oN_stride + tile_hbytes; size_t bh = block_height; for (; bh >= 8; bh -= 8) { const uint8x8_t v3_0 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_1 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_2 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_3 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_4 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_5 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_6 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_7 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); o = (uint8_t*) ((uintptr_t) o + oN_offset); vst1_u8(o, v0_3.val[1]); if XNN_UNPREDICTABLE(block_width > 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_3.val[0]); if XNN_UNPREDICTABLE(block_width >= 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_2.val[1]); if XNN_UNPREDICTABLE(block_width > 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_2.val[0]); if XNN_UNPREDICTABLE(block_width >= 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_1.val[1]); if XNN_UNPREDICTABLE(block_width > 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_1.val[0]); if XNN_UNPREDICTABLE(block_width >= 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_0.val[1]); if XNN_UNPREDICTABLE(block_width > 1) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_u8(o, v0_0.val[0]); } o = (uint8_t*) ((uintptr_t) o + tile_hbytes); if (bh != 0) { const uint8x8_t v3_0 = vld1_u8(i0); const uint8_t *i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const uint8x8_t v3_1 = vld1_u8(i1); const uint8_t *i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i1; } const uint8x8_t v3_2 = vld1_u8(i2); const uint8_t *i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); if XNN_UNPREDICTABLE(bh < 4) { i3 = i2; } const uint8x8_t v3_3 = vld1_u8(i3); const uint8_t *i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); if XNN_UNPREDICTABLE(bh <= 4) { i4 = i3; } const uint8x8_t v3_4 = vld1_u8(i4); const uint8_t *i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); if XNN_UNPREDICTABLE(bh < 6) { i5 = i4; } const uint8x8_t v3_5 = vld1_u8(i5); const uint8_t *i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); if XNN_UNPREDICTABLE(bh <= 6) { i6 = i5; } const uint8x8_t v3_6 = vld1_u8(i6); const uint8x8_t v3_7 = vmov_n_u8(0); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); uint8x8_t v0_low = v0_0.val[0]; uint8x8_t v1_low = v0_0.val[1]; uint8x8_t v2_low = v0_1.val[0]; uint8x8_t v3_low = v0_1.val[1]; uint8x8_t v4_low = v0_2.val[0]; uint8x8_t v5_low = v0_2.val[1]; uint8x8_t v6_low = v0_3.val[0]; uint8x8_t v7_low = v0_3.val[1]; if (bh & 4) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u32((void*) o, vreinterpret_u32_u8(v7_low), 0); if XNN_UNPREDICTABLE(block_width > 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v6_low), 0); if XNN_UNPREDICTABLE(block_width >= 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v5_low), 0); if XNN_UNPREDICTABLE(block_width > 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v4_low), 0); if XNN_UNPREDICTABLE(block_width >= 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v3_low), 0); if XNN_UNPREDICTABLE(block_width > 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v2_low), 0); if XNN_UNPREDICTABLE(block_width >= 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v1_low), 0); if XNN_UNPREDICTABLE(block_width > 1) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v0_low), 0); o += 4; v0_low = vext_u8(v0_low, v0_low, 4); v1_low = vext_u8(v1_low, v1_low, 4); v2_low = vext_u8(v2_low, v2_low, 4); v3_low = vext_u8(v3_low, v3_low, 4); v4_low = vext_u8(v4_low, v4_low, 4); v5_low = vext_u8(v5_low, v5_low, 4); v6_low = vext_u8(v6_low, v6_low, 4); v7_low = vext_u8(v7_low, v7_low, 4); } if (bh & 2) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u16((void*) o, vreinterpret_u16_u8(v7_low), 0); if XNN_UNPREDICTABLE(block_width > 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v6_low), 0); if XNN_UNPREDICTABLE(block_width >= 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v5_low), 0); if XNN_UNPREDICTABLE(block_width > 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v4_low), 0); if XNN_UNPREDICTABLE(block_width >= 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v3_low), 0); if XNN_UNPREDICTABLE(block_width > 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v2_low), 0); if XNN_UNPREDICTABLE(block_width >= 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v1_low), 0); if XNN_UNPREDICTABLE(block_width > 1) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v0_low), 0); o += 2; v0_low = vext_u8(v0_low, v0_low, 2); v1_low = vext_u8(v1_low, v1_low, 2); v2_low = vext_u8(v2_low, v2_low, 2); v3_low = vext_u8(v3_low, v3_low, 2); v4_low = vext_u8(v4_low, v4_low, 2); v5_low = vext_u8(v5_low, v5_low, 2); v6_low = vext_u8(v6_low, v6_low, 2); v7_low = vext_u8(v7_low, v7_low, 2); } if (bh & 1) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u8(o, v7_low, 0); if XNN_UNPREDICTABLE(block_width > 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v6_low, 0); if XNN_UNPREDICTABLE(block_width >= 7) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v5_low, 0); if XNN_UNPREDICTABLE(block_width > 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v4_low, 0); if XNN_UNPREDICTABLE(block_width >= 5) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v3_low, 0); if XNN_UNPREDICTABLE(block_width > 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v2_low, 0); if XNN_UNPREDICTABLE(block_width >= 3) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v1_low, 0); if XNN_UNPREDICTABLE(block_width > 1) { o = (uint8_t*) ((uintptr_t) o + minus_output_stride); } vst1_lane_u8(o, v0_low, 0); } } i0 = (const uint8_t*) ((uintptr_t) i0 + input_reset); o = (uint8_t*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
11,553
40.412186
123
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-8x8-reuse-mov-zip-neon.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/neon-zip.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__8x8_reuse_mov_zip_neon( const uint8_t* input, uint8_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(uint8_t)); assert(input_stride >= block_width * sizeof(uint8_t)); const size_t tile_height = 8; const size_t tile_width = 8; const size_t tile_hbytes = tile_height * sizeof(uint8_t); const size_t tile_wbytes = tile_width * sizeof(uint8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(uint8_t) - tile_hbytes; const uint8_t* i0 = input; uint8_t* o = (uint8_t*) ((uintptr_t) output - tile_hbytes); const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 7); const size_t oN_stride = rem * output_stride; const size_t oN_offset = oN_stride + tile_hbytes; size_t bh = block_height; for (; bh >= 8; bh -= 8) { const uint8x8_t v3_0 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_1 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_2 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_3 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_4 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_5 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_6 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_7 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); o = (uint8_t*) ((uintptr_t) o + oN_offset); vst1_u8(o, v0_3.val[1]); uint8_t *oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 7) { o = oN; } vst1_u8(o, v0_3.val[0]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 7) { o = oN; } vst1_u8(o, v0_2.val[1]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 5) { o = oN; } vst1_u8(o, v0_2.val[0]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 5) { o = oN; } vst1_u8(o, v0_1.val[1]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } vst1_u8(o, v0_1.val[0]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } vst1_u8(o, v0_0.val[1]); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } vst1_u8(o, v0_0.val[0]); } o = (uint8_t*) ((uintptr_t) o + tile_hbytes); if (bh != 0) { const uint8x8_t v3_0 = vld1_u8(i0); const uint8_t *i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const uint8x8_t v3_1 = vld1_u8(i1); const uint8_t *i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i1; } const uint8x8_t v3_2 = vld1_u8(i2); const uint8_t *i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); if XNN_UNPREDICTABLE(bh < 4) { i3 = i2; } const uint8x8_t v3_3 = vld1_u8(i3); const uint8_t *i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); if XNN_UNPREDICTABLE(bh <= 4) { i4 = i3; } const uint8x8_t v3_4 = vld1_u8(i4); const uint8_t *i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); if XNN_UNPREDICTABLE(bh < 6) { i5 = i4; } const uint8x8_t v3_5 = vld1_u8(i5); const uint8_t *i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); if XNN_UNPREDICTABLE(bh <= 6) { i6 = i5; } const uint8x8_t v3_6 = vld1_u8(i6); const uint8x8_t v3_7 = vmov_n_u8(0); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); uint8x8_t v0_low = v0_0.val[0]; uint8x8_t v1_low = v0_0.val[1]; uint8x8_t v2_low = v0_1.val[0]; uint8x8_t v3_low = v0_1.val[1]; uint8x8_t v4_low = v0_2.val[0]; uint8x8_t v5_low = v0_2.val[1]; uint8x8_t v6_low = v0_3.val[0]; uint8x8_t v7_low = v0_3.val[1]; if (bh & 4) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u32((void*) o, vreinterpret_u32_u8(v7_low), 0); uint8_t *oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 7) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 7) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 5) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 5) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v1_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } vst1_lane_u32((void*) o, vreinterpret_u32_u8(v0_low), 0); o += 4; v0_low = vext_u8(v0_low, v0_low, 4); v1_low = vext_u8(v1_low, v1_low, 4); v2_low = vext_u8(v2_low, v2_low, 4); v3_low = vext_u8(v3_low, v3_low, 4); v4_low = vext_u8(v4_low, v4_low, 4); v5_low = vext_u8(v5_low, v5_low, 4); v6_low = vext_u8(v6_low, v6_low, 4); v7_low = vext_u8(v7_low, v7_low, 4); } if (bh & 2) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u16((void*) o, vreinterpret_u16_u8(v7_low), 0); uint8_t *oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 7) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 7) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 5) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 5) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v1_low), 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } vst1_lane_u16((void*) o, vreinterpret_u16_u8(v0_low), 0); o += 2; v0_low = vext_u8(v0_low, v0_low, 2); v1_low = vext_u8(v1_low, v1_low, 2); v2_low = vext_u8(v2_low, v2_low, 2); v3_low = vext_u8(v3_low, v3_low, 2); v4_low = vext_u8(v4_low, v4_low, 2); v5_low = vext_u8(v5_low, v5_low, 2); v6_low = vext_u8(v6_low, v6_low, 2); v7_low = vext_u8(v7_low, v7_low, 2); } if (bh & 1) { o = (uint8_t*) ((uintptr_t) o + oN_stride); vst1_lane_u8(o, v7_low, 0); uint8_t *oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 7) { o = oN; } vst1_lane_u8(o, v6_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 7) { o = oN; } vst1_lane_u8(o, v5_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 5) { o = oN; } vst1_lane_u8(o, v4_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 5) { o = oN; } vst1_lane_u8(o, v3_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 3) { o = oN; } vst1_lane_u8(o, v2_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width >= 3) { o = oN; } vst1_lane_u8(o, v1_low, 0); oN = (uint8_t*) ((uintptr_t) o + minus_output_stride); if XNN_UNPREDICTABLE(block_width > 1) { o = oN; } vst1_lane_u8(o, v0_low, 0); } } i0 = (const uint8_t*) ((uintptr_t) i0 + input_reset); o = (uint8_t*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
12,051
38.257329
123
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-8x8-reuse-multi-zip-neon.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/neon-zip.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__8x8_reuse_multi_zip_neon( const uint8_t* input, uint8_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(uint8_t)); assert(input_stride >= block_width * sizeof(uint8_t)); const size_t tile_height = 8; const size_t tile_width = 8; const size_t tile_hbytes = tile_height * sizeof(uint8_t); const size_t tile_wbytes = tile_width * sizeof(uint8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(uint8_t); const uint8_t* i0 = input; uint8_t* o0 = (uint8_t*) output; uint8_t* o1 = (uint8_t*) ((uintptr_t) o0 + output_stride); uint8_t* o2 = (uint8_t*) ((uintptr_t) o1 + output_stride); uint8_t* o3 = (uint8_t*) ((uintptr_t) o2 + output_stride); uint8_t* o4 = (uint8_t*) ((uintptr_t) o3 + output_stride); uint8_t* o5 = (uint8_t*) ((uintptr_t) o4 + output_stride); uint8_t* o6 = (uint8_t*) ((uintptr_t) o5 + output_stride); uint8_t* o7 = (uint8_t*) ((uintptr_t) o6 + output_stride); do { if XNN_UNPREDICTABLE(block_width < 2) { o1 = o0; } if XNN_UNPREDICTABLE(block_width <= 2) { o2 = o0; } if XNN_UNPREDICTABLE(block_width < 4) { o3 = o0; } if XNN_UNPREDICTABLE(block_width <= 4) { o4 = o0; } if XNN_UNPREDICTABLE(block_width < 6) { o5 = o0; } if XNN_UNPREDICTABLE(block_width <= 6) { o6 = o0; } if XNN_UNPREDICTABLE(block_width < 8) { o7 = o0; } size_t bh = block_height; for (; bh >= 8; bh -= 8) { const uint8x8_t v3_0 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_1 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_2 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_3 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_4 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_5 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_6 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_7 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); vst1_u8(o7, v0_3.val[1]); o7 = (uint8_t*) ((uintptr_t) o7 + tile_hbytes); vst1_u8(o6, v0_3.val[0]); o6 = (uint8_t*) ((uintptr_t) o6 + tile_hbytes); vst1_u8(o5, v0_2.val[1]); o5 = (uint8_t*) ((uintptr_t) o5 + tile_hbytes); vst1_u8(o4, v0_2.val[0]); o4 = (uint8_t*) ((uintptr_t) o4 + tile_hbytes); vst1_u8(o3, v0_1.val[1]); o3 = (uint8_t*) ((uintptr_t) o3 + tile_hbytes); vst1_u8(o2, v0_1.val[0]); o2 = (uint8_t*) ((uintptr_t) o2 + tile_hbytes); vst1_u8(o1, v0_0.val[1]); o1 = (uint8_t*) ((uintptr_t) o1 + tile_hbytes); vst1_u8(o0, v0_0.val[0]); o0 = (uint8_t*) ((uintptr_t) o0 + tile_hbytes); } if (bh != 0) { const uint8x8_t v3_0 = vld1_u8(i0); const uint8_t *i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const uint8x8_t v3_1 = vld1_u8(i1); const uint8_t *i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i1; } const uint8x8_t v3_2 = vld1_u8(i2); const uint8_t *i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); if XNN_UNPREDICTABLE(bh < 4) { i3 = i2; } const uint8x8_t v3_3 = vld1_u8(i3); const uint8_t *i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); if XNN_UNPREDICTABLE(bh <= 4) { i4 = i3; } const uint8x8_t v3_4 = vld1_u8(i4); const uint8_t *i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); if XNN_UNPREDICTABLE(bh < 6) { i5 = i4; } const uint8x8_t v3_5 = vld1_u8(i5); const uint8_t *i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); if XNN_UNPREDICTABLE(bh <= 6) { i6 = i5; } const uint8x8_t v3_6 = vld1_u8(i6); const uint8x8_t v3_7 = vmov_n_u8(0); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); uint8x8_t v0_low = v0_0.val[0]; uint8x8_t v1_low = v0_0.val[1]; uint8x8_t v2_low = v0_1.val[0]; uint8x8_t v3_low = v0_1.val[1]; uint8x8_t v4_low = v0_2.val[0]; uint8x8_t v5_low = v0_2.val[1]; uint8x8_t v6_low = v0_3.val[0]; uint8x8_t v7_low = v0_3.val[1]; if (bh & 4) { vst1_lane_u32((void*) o7, vreinterpret_u32_u8(v7_low), 0); o7 += 4; vst1_lane_u32((void*) o6, vreinterpret_u32_u8(v6_low), 0); o6 += 4; vst1_lane_u32((void*) o5, vreinterpret_u32_u8(v5_low), 0); o5 += 4; vst1_lane_u32((void*) o4, vreinterpret_u32_u8(v4_low), 0); o4 += 4; vst1_lane_u32((void*) o3, vreinterpret_u32_u8(v3_low), 0); o3 += 4; vst1_lane_u32((void*) o2, vreinterpret_u32_u8(v2_low), 0); o2 += 4; vst1_lane_u32((void*) o1, vreinterpret_u32_u8(v1_low), 0); o1 += 4; vst1_lane_u32((void*) o0, vreinterpret_u32_u8(v0_low), 0); o0 += 4; v0_low = vext_u8(v0_low, v0_low, 4); v1_low = vext_u8(v1_low, v1_low, 4); v2_low = vext_u8(v2_low, v2_low, 4); v3_low = vext_u8(v3_low, v3_low, 4); v4_low = vext_u8(v4_low, v4_low, 4); v5_low = vext_u8(v5_low, v5_low, 4); v6_low = vext_u8(v6_low, v6_low, 4); v7_low = vext_u8(v7_low, v7_low, 4); } if (bh & 2) { vst1_lane_u16((void*) o7, vreinterpret_u16_u8(v7_low), 0); o7 += 2; vst1_lane_u16((void*) o6, vreinterpret_u16_u8(v6_low), 0); o6 += 2; vst1_lane_u16((void*) o5, vreinterpret_u16_u8(v5_low), 0); o5 += 2; vst1_lane_u16((void*) o4, vreinterpret_u16_u8(v4_low), 0); o4 += 2; vst1_lane_u16((void*) o3, vreinterpret_u16_u8(v3_low), 0); o3 += 2; vst1_lane_u16((void*) o2, vreinterpret_u16_u8(v2_low), 0); o2 += 2; vst1_lane_u16((void*) o1, vreinterpret_u16_u8(v1_low), 0); o1 += 2; vst1_lane_u16((void*) o0, vreinterpret_u16_u8(v0_low), 0); o0 += 2; v0_low = vext_u8(v0_low, v0_low, 2); v1_low = vext_u8(v1_low, v1_low, 2); v2_low = vext_u8(v2_low, v2_low, 2); v3_low = vext_u8(v3_low, v3_low, 2); v4_low = vext_u8(v4_low, v4_low, 2); v5_low = vext_u8(v5_low, v5_low, 2); v6_low = vext_u8(v6_low, v6_low, 2); v7_low = vext_u8(v7_low, v7_low, 2); } if (bh & 1) { vst1_lane_u8(o7, v7_low, 0); vst1_lane_u8(o6, v6_low, 0); vst1_lane_u8(o5, v5_low, 0); vst1_lane_u8(o4, v4_low, 0); vst1_lane_u8(o3, v3_low, 0); vst1_lane_u8(o2, v2_low, 0); vst1_lane_u8(o1, v1_low, 0); vst1_lane_u8(o0, v0_low, 0); } } i0 = (const uint8_t*) ((uintptr_t) i0 + input_reset); o0 = (uint8_t*) ((uintptr_t) o0 + output_reset); o1 = (uint8_t*) ((uintptr_t) o1 + output_reset); o2 = (uint8_t*) ((uintptr_t) o2 + output_reset); o3 = (uint8_t*) ((uintptr_t) o3 + output_reset); o4 = (uint8_t*) ((uintptr_t) o4 + output_reset); o5 = (uint8_t*) ((uintptr_t) o5 + output_reset); o6 = (uint8_t*) ((uintptr_t) o6 + output_reset); o7 = (uint8_t*) ((uintptr_t) o7 + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
9,469
41.850679
109
c
XNNPACK
XNNPACK-master/src/x8-transposec/gen/x8-transposec-8x8-reuse-switch-zip-neon.c
// Auto-generated file. Do not edit! // Template: src/x32-transposec/neon-zip.c.in // Generator: tools/xngen // // Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <assert.h> #include <xnnpack/common.h> #include <xnnpack/math.h> #include <xnnpack/transpose.h> void xnn_x8_transposec_ukernel__8x8_reuse_switch_zip_neon( const uint8_t* input, uint8_t* output, size_t input_stride, size_t output_stride, size_t block_width, size_t block_height, const union xnn_x8_transpose_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_OOB_READS { assert(output_stride >= block_height * sizeof(uint8_t)); assert(input_stride >= block_width * sizeof(uint8_t)); const size_t tile_height = 8; const size_t tile_width = 8; const size_t tile_hbytes = tile_height * sizeof(uint8_t); const size_t tile_wbytes = tile_width * sizeof(uint8_t); const size_t input_reset = tile_wbytes - round_down_po2(block_height, tile_height) * input_stride; const size_t output_reset = tile_width * output_stride - round_down_po2(block_height, 2) * sizeof(uint8_t); const uint8_t* i0 = input; uint8_t* o = (uint8_t*) output; const size_t minus_output_stride = -output_stride; do { const size_t rem = min(block_width - 1, 7); const size_t oN_stride = rem * output_stride; size_t bh = block_height; for (; bh >= 8; bh -= 8) { const uint8x8_t v3_0 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_1 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_2 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_3 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_4 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_5 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_6 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8_t v3_7 = vld1_u8(i0); i0 = (uint8_t*) ((uintptr_t) i0 + input_stride); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); uint8_t *oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 7: vst1_u8(oN, v0_3.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_u8(oN, v0_3.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_u8(oN, v0_2.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_u8(oN, v0_2.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_u8(oN, v0_1.val[1]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_u8(oN, v0_1.val[0]); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_u8(oN, v0_0.val[1]); case 0: vst1_u8(o, v0_0.val[0]); o = (uint8_t*) ((uintptr_t) o + tile_hbytes); break; default: XNN_UNREACHABLE; } } if (bh != 0) { const uint8x8_t v3_0 = vld1_u8(i0); const uint8_t *i1 = (const uint8_t*) ((uintptr_t) i0 + input_stride); if XNN_UNPREDICTABLE(bh < 2) { i1 = i0; } const uint8x8_t v3_1 = vld1_u8(i1); const uint8_t *i2 = (const uint8_t*) ((uintptr_t) i1 + input_stride); if XNN_UNPREDICTABLE(bh <= 2) { i2 = i1; } const uint8x8_t v3_2 = vld1_u8(i2); const uint8_t *i3 = (const uint8_t*) ((uintptr_t) i2 + input_stride); if XNN_UNPREDICTABLE(bh < 4) { i3 = i2; } const uint8x8_t v3_3 = vld1_u8(i3); const uint8_t *i4 = (const uint8_t*) ((uintptr_t) i3 + input_stride); if XNN_UNPREDICTABLE(bh <= 4) { i4 = i3; } const uint8x8_t v3_4 = vld1_u8(i4); const uint8_t *i5 = (const uint8_t*) ((uintptr_t) i4 + input_stride); if XNN_UNPREDICTABLE(bh < 6) { i5 = i4; } const uint8x8_t v3_5 = vld1_u8(i5); const uint8_t *i6 = (const uint8_t*) ((uintptr_t) i5 + input_stride); if XNN_UNPREDICTABLE(bh <= 6) { i6 = i5; } const uint8x8_t v3_6 = vld1_u8(i6); const uint8x8_t v3_7 = vmov_n_u8(0); const uint8x8x2_t v2_0 = vzip_u8(v3_0, v3_4); const uint8x8x2_t v2_1 = vzip_u8(v3_1, v3_5); const uint8x8x2_t v2_2 = vzip_u8(v3_2, v3_6); const uint8x8x2_t v2_3 = vzip_u8(v3_3, v3_7); const uint8x8x2_t v1_0 = vzip_u8(v2_0.val[0], v2_2.val[0]); const uint8x8x2_t v1_1 = vzip_u8(v2_0.val[1], v2_2.val[1]); const uint8x8x2_t v1_2 = vzip_u8(v2_1.val[0], v2_3.val[0]); const uint8x8x2_t v1_3 = vzip_u8(v2_1.val[1], v2_3.val[1]); const uint8x8x2_t v0_0 = vzip_u8(v1_0.val[0], v1_2.val[0]); const uint8x8x2_t v0_1 = vzip_u8(v1_0.val[1], v1_2.val[1]); const uint8x8x2_t v0_2 = vzip_u8(v1_1.val[0], v1_3.val[0]); const uint8x8x2_t v0_3 = vzip_u8(v1_1.val[1], v1_3.val[1]); uint8x8_t v0_low = v0_0.val[0]; uint8x8_t v1_low = v0_0.val[1]; uint8x8_t v2_low = v0_1.val[0]; uint8x8_t v3_low = v0_1.val[1]; uint8x8_t v4_low = v0_2.val[0]; uint8x8_t v5_low = v0_2.val[1]; uint8x8_t v6_low = v0_3.val[0]; uint8x8_t v7_low = v0_3.val[1]; if (bh & 4) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 7: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v7_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_lane_u32((void*) oN, vreinterpret_u32_u8(v1_low), 0); case 0: vst1_lane_u32((void*) o, vreinterpret_u32_u8(v0_low), 0); o += 4; break; default: XNN_UNREACHABLE; } v0_low = vext_u8(v0_low, v0_low, 4); v1_low = vext_u8(v1_low, v1_low, 4); v2_low = vext_u8(v2_low, v2_low, 4); v3_low = vext_u8(v3_low, v3_low, 4); v4_low = vext_u8(v4_low, v4_low, 4); v5_low = vext_u8(v5_low, v5_low, 4); v6_low = vext_u8(v6_low, v6_low, 4); v7_low = vext_u8(v7_low, v7_low, 4); } if (bh & 2) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 7: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v7_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v6_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v5_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v4_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v3_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v2_low), 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_lane_u16((void*) oN, vreinterpret_u16_u8(v1_low), 0); case 0: vst1_lane_u16((void*) o, vreinterpret_u16_u8(v0_low), 0); o += 2; break; default: XNN_UNREACHABLE; } v0_low = vext_u8(v0_low, v0_low, 2); v1_low = vext_u8(v1_low, v1_low, 2); v2_low = vext_u8(v2_low, v2_low, 2); v3_low = vext_u8(v3_low, v3_low, 2); v4_low = vext_u8(v4_low, v4_low, 2); v5_low = vext_u8(v5_low, v5_low, 2); v6_low = vext_u8(v6_low, v6_low, 2); v7_low = vext_u8(v7_low, v7_low, 2); } if (bh & 1) { uint8_t* oN = (uint8_t*) ((uintptr_t) o + oN_stride); switch (rem) { case 7: vst1_lane_u8(oN, v7_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 6: vst1_lane_u8(oN, v6_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 5: vst1_lane_u8(oN, v5_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 4: vst1_lane_u8(oN, v4_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 3: vst1_lane_u8(oN, v3_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 2: vst1_lane_u8(oN, v2_low, 0); oN = (uint8_t*) ((uintptr_t) oN + minus_output_stride); case 1: vst1_lane_u8(oN, v1_low, 0); case 0: vst1_lane_u8(o, v0_low, 0); break; default: XNN_UNREACHABLE; } } } i0 = (const uint8_t*) ((uintptr_t) i0 + input_reset); o = (uint8_t*) ((uintptr_t) o + output_reset); block_width = doz(block_width, tile_width); } while (block_width != 0); }
10,560
42.106122
126
c
XNNPACK
XNNPACK-master/src/x8-zip/x8-zip-x2-neon.c
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <xnnpack/zip.h> void xnn_x8_zip_x2_ukernel__neon( size_t n, const uint8_t* input, uint8_t* output) { const uint8_t* x = input; const uint8_t* y = (const uint8_t*) ((uintptr_t) x + n); uint8_t* o = output; if (n >= 8) { do { uint8x8x2_t vxy; vxy.val[0] = vld1_u8(x); x += 8; vxy.val[1] = vld1_u8(y); y += 8; vst2_u8(o, vxy); o += 16;; n -= 8; } while (n >= 8); if (n != 0) { const size_t address_increment = n - 8; uint8x8x2_t vxy; vxy.val[0] = vld1_u8((const uint8_t*) ((uintptr_t) x + address_increment)); vxy.val[1] = vld1_u8((const uint8_t*) ((uintptr_t) y + address_increment)); vst2_u8((uint8_t*) ((uintptr_t) o + address_increment * 2), vxy); } } else { do { const uint8_t vx = *x++; const uint8_t vy = *y++; o[0] = vx; o[1] = vy; o += 2; } while (--n != 0); } }
1,186
23.729167
81
c
XNNPACK
XNNPACK-master/src/x8-zip/x8-zip-x2-sse2.c
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <emmintrin.h> #include <xnnpack/zip.h> void xnn_x8_zip_x2_ukernel__sse2( size_t n, const uint8_t* input, uint8_t* output) { const uint8_t* x = input; const uint8_t* y = (const uint8_t*) ((uintptr_t) x + n); uint8_t* o = output; if (n >= 16) { do { const __m128i vx = _mm_loadu_si128((const __m128i*) x); x += 16; const __m128i vy = _mm_loadu_si128((const __m128i*) y); y += 16; const __m128i vxy_lo = _mm_unpacklo_epi8(vx, vy); const __m128i vxy_hi = _mm_unpackhi_epi8(vx, vy); _mm_storeu_si128((__m128i*) o, vxy_lo); _mm_storeu_si128((__m128i*) (o + 16), vxy_hi); o = (void*) ((uintptr_t) o + 32); n -= 16; } while (n >= 16); if (n != 0) { const size_t address_increment = n - 16; const __m128i vx = _mm_loadu_si128((const __m128i*) ((uintptr_t) x + address_increment)); const __m128i vy = _mm_loadu_si128((const __m128i*) ((uintptr_t) y + address_increment)); const __m128i vxy_lo = _mm_unpacklo_epi8(vx, vy); const __m128i vxy_hi = _mm_unpackhi_epi8(vx, vy); o = (void*) ((uintptr_t) o + address_increment * 2); _mm_storeu_si128((__m128i*) o, vxy_lo); _mm_storeu_si128((__m128i*) o + 1, vxy_hi); } } else { do { const uint8_t vx = *x++; const uint8_t vy = *y++; o[0] = vx; o[1] = vy; o += 2; } while (--n != 0); } }
1,662
28.696429
95
c
XNNPACK
XNNPACK-master/src/x8-zip/x8-zip-x3-neon.c
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <xnnpack/zip.h> void xnn_x8_zip_x3_ukernel__neon( size_t n, const uint8_t* input, uint8_t* output) { const uint8_t* x = input; const uint8_t* y = (const uint8_t*) ((uintptr_t) x + n); const uint8_t* z = (const uint8_t*) ((uintptr_t) y + n); uint8_t* o = output; if (n >= 8) { do { uint8x8x3_t vxyz; vxyz.val[0] = vld1_u8(x); x += 8; vxyz.val[1] = vld1_u8(y); y += 8; vxyz.val[2] = vld1_u8(z); z += 8; vst3_u8(o, vxyz); o += 24; n -= 8; } while (n >= 8); if (n != 0) { const size_t address_increment = n - 8; uint8x8x3_t vxyz; vxyz.val[0] = vld1_u8((const uint8_t*) ((uintptr_t) x + address_increment)); vxyz.val[1] = vld1_u8((const uint8_t*) ((uintptr_t) y + address_increment)); vxyz.val[2] = vld1_u8((const uint8_t*) ((uintptr_t) z + address_increment)); vst3_u8((uint8_t*) ((uintptr_t) o + address_increment * 3), vxyz); } } else { do { const uint8_t vx = *x++; const uint8_t vy = *y++; const uint8_t vz = *z++; o[0] = vx; o[1] = vy; o[2] = vz; o += 3; } while (--n != 0); } }
1,423
25.867925
82
c
XNNPACK
XNNPACK-master/src/x8-zip/x8-zip-x3-sse2.c
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <emmintrin.h> #include <xnnpack/zip.h> void xnn_x8_zip_x3_ukernel__sse2( size_t n, const uint8_t* input, uint8_t* output) { const uint8_t* x = input; const uint8_t* y = (const uint8_t*) ((uintptr_t) x + n); const uint8_t* z = (const uint8_t*) ((uintptr_t) y + n); uint8_t* o = output; if (n >= 16) { const __m128i vmask0x00FF00FF = _mm_set1_epi16(0x00FF); const __m128i vmask0x0000FFFF = _mm_set1_epi32(0x0000FFFF); do { // vx = ( x15, x14, x13, x12, x11, x10, x9, x8, x7, x6, x5, x4, x3, x2, x1, x0 ) const __m128i vx = _mm_loadu_si128((const __m128i*) x); x += 16; // vy = ( y15, y14, y13, y12, y11, y10, y9, y8, y7, y6, y5, y4, y3, y2, y1, y0 ) const __m128i vy = _mm_loadu_si128((const __m128i*) y); y += 16; // vz = ( z15, z14, z13, z12, z11, z10, z9, z8, z7, z6, z5, z4, z3, z2, z1, z0 ) const __m128i vz = _mm_loadu_si128((const __m128i*) z); z += 16; // vxeye = ( y14, x14, y12, x12, y10, x10, y8, x8, y6, x6, y4, x4, y2, x2, y0, x0 ) const __m128i vxeye = _mm_or_si128(_mm_and_si128(vx, vmask0x00FF00FF), _mm_slli_epi16(vy, 8)); // vyozo = ( z15, y15, z13, y13, z11, y11, z9, y9, z7, y7, z5, y5, z3, y3, z1, y1 ) const __m128i vyozo = _mm_or_si128(_mm_andnot_si128(vmask0x00FF00FF, vz), _mm_srli_epi16(vy, 8)); // vzoxo = ( x15, z14, x13, z12, x11, z10, x9, z8, x7, z6, x5, z4, x3, z2, x1, z0 ) const __m128i vzexo = _mm_or_si128(_mm_and_si128(vz, vmask0x00FF00FF), _mm_andnot_si128(vmask0x00FF00FF, vx)); // vxeyezexo = ( x13, z12, y12, x12, x9, z8, y8, x8, x5, z4, y4, x4, x1, z0, y0, x0 ) const __m128i vxeyezexo = _mm_or_si128(_mm_and_si128(vxeye, vmask0x0000FFFF), _mm_slli_epi32(vzexo, 16)); // vyozoxeye = ( y14, x14, z13, y13, y10, x10, z9, y9, y6, x6, z5, y5, y2, x2, z1, y1 ) const __m128i vyozoxeye = _mm_or_si128(_mm_and_si128(vyozo, vmask0x0000FFFF), _mm_andnot_si128(vmask0x0000FFFF, vxeye)); // vzexoyozo = ( z15, y15, x15, z14, z11, y11, x11, z10, z7, y7, x7, z6, z3, y3, x3, z2 ) const __m128i vzexoyozo = _mm_or_si128(_mm_andnot_si128(vmask0x0000FFFF, vyozo), _mm_srli_epi32(vzexo, 16)); // vtemp0 = ( x13, z12, y12, x12, x5, z4, y4, x4, z11, y11, x11, z10, z3, y3, x3, z2 ) const __m128i vtemp0 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vzexoyozo), _mm_castsi128_ps(vxeyezexo), _MM_SHUFFLE(3, 1, 2, 0))); // vtemp1 = ( y10, x10, z9, y9, y2, x2, z1, y1, x9, z8, y8, x8, x1, z0, y0, x0 ) const __m128i vtemp1 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vxeyezexo), _mm_castsi128_ps(vyozoxeye), _MM_SHUFFLE(2, 0, 2, 0))); // vtemp2 = ( z15, y15, x15, z14, z7, y7, x7, z6, y14, x14, z13, y13, y6, x6, z5, y5 ) const __m128i vtemp2 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vyozoxeye), _mm_castsi128_ps(vzexoyozo), _MM_SHUFFLE(3, 1, 3, 1))); // vxyz0 = ( x5, z4, y4, x4, z3, y3, x3, z2, y2, x2, z1, y1, x1, z0, y0, x0 ) const __m128i vxyz0 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vtemp1), _mm_castsi128_ps(vtemp0), _MM_SHUFFLE(2, 0, 2, 0))); // vxyz1 = ( y10, x10, z9, y9, x9, z8, y8, x8, z7, y7, x7, z6, y6, x6, z5, y5 ) const __m128i vxyz1 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vtemp2), _mm_castsi128_ps(vtemp1), _MM_SHUFFLE(3, 1, 2, 0))); // vxyz2 = ( z15, y15, x15, z14, y14, x14, z13, y13, x13, z12, y12, x12, z11, y11, x11, z10 ) const __m128i vxyz2 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vtemp0), _mm_castsi128_ps(vtemp2), _MM_SHUFFLE(3, 1, 3, 1))); _mm_storeu_si128((__m128i*) o, vxyz0); _mm_storeu_si128((__m128i*) o + 1, vxyz1); _mm_storeu_si128((__m128i*) o + 2, vxyz2); o += 48; n -= 16; } while (n >= 16); if (n != 0) { const size_t address_increment = n - 16; // vx = ( x15, x14, x13, x12, x11, x10, x9, x8, x7, x6, x5, x4, x3, x2, x1, x0 ) const __m128i vx = _mm_loadu_si128((const __m128i*) ((uintptr_t) x + address_increment)); // vy = ( y15, y14, y13, y12, y11, y10, y9, y8, y7, y6, y5, y4, y3, y2, y1, y0 ) const __m128i vy = _mm_loadu_si128((const __m128i*) ((uintptr_t) y + address_increment)); // vz = ( z15, z14, z13, z12, z11, z10, z9, z8, z7, z6, z5, z4, z3, z2, z1, z0 ) const __m128i vz = _mm_loadu_si128((const __m128i*) ((uintptr_t) z + address_increment)); // vxeye = ( y14, x14, y12, x12, y10, x10, y8, x8, y6, x6, y4, x4, y2, x2, y0, x0 ) const __m128i vxeye = _mm_or_si128(_mm_and_si128(vx, vmask0x00FF00FF), _mm_slli_epi16(vy, 8)); // vyozo = ( z15, y15, z13, y13, z11, y11, z9, y9, z7, y7, z5, y5, z3, y3, z1, y1 ) const __m128i vyozo = _mm_or_si128(_mm_andnot_si128(vmask0x00FF00FF, vz), _mm_srli_epi16(vy, 8)); // vzoxo = ( x15, z14, x13, z12, x11, z10, x9, z8, x7, z6, x5, z4, x3, z2, x1, z0 ) const __m128i vzexo = _mm_or_si128(_mm_and_si128(vz, vmask0x00FF00FF), _mm_andnot_si128(vmask0x00FF00FF, vx)); // vxeyezexo = ( x13, z12, y12, x12, x9, z8, y8, x8, x5, z4, y4, x4, x1, z0, y0, x0 ) const __m128i vxeyezexo = _mm_or_si128(_mm_and_si128(vxeye, vmask0x0000FFFF), _mm_slli_epi32(vzexo, 16)); // vyozoxeye = ( y14, x14, z13, y13, y10, x10, z9, y9, y6, x6, z5, y5, y2, x2, z1, y1 ) const __m128i vyozoxeye = _mm_or_si128(_mm_and_si128(vyozo, vmask0x0000FFFF), _mm_andnot_si128(vmask0x0000FFFF, vxeye)); // vzexoyozo = ( z15, y15, x15, z14, z11, y11, x11, z10, z7, y7, x7, z6, z3, y3, x3, z2 ) const __m128i vzexoyozo = _mm_or_si128(_mm_andnot_si128(vmask0x0000FFFF, vyozo), _mm_srli_epi32(vzexo, 16)); // vtemp0 = ( x13, z12, y12, x12, x5, z4, y4, x4, z11, y11, x11, z10, z3, y3, x3, z2 ) const __m128i vtemp0 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vzexoyozo), _mm_castsi128_ps(vxeyezexo), _MM_SHUFFLE(3, 1, 2, 0))); // vtemp1 = ( y10, x10, z9, y9, y2, x2, z1, y1, x9, z8, y8, x8, x1, z0, y0, x0 ) const __m128i vtemp1 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vxeyezexo), _mm_castsi128_ps(vyozoxeye), _MM_SHUFFLE(2, 0, 2, 0))); // vtemp2 = ( z15, y15, x15, z14, z7, y7, x7, z6, y14, x14, z13, y13, y6, x6, z5, y5 ) const __m128i vtemp2 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vyozoxeye), _mm_castsi128_ps(vzexoyozo), _MM_SHUFFLE(3, 1, 3, 1))); // vxyz0 = ( x5, z4, y4, x4, z3, y3, x3, z2, y2, x2, z1, y1, x1, z0, y0, x0 ) const __m128i vxyz0 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vtemp1), _mm_castsi128_ps(vtemp0), _MM_SHUFFLE(2, 0, 2, 0))); // vxyz1 = ( y10, x10, z9, y9, x9, z8, y8, x8, z7, y7, x7, z6, y6, x6, z5, y5 ) const __m128i vxyz1 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vtemp2), _mm_castsi128_ps(vtemp1), _MM_SHUFFLE(3, 1, 2, 0))); // vxyz2 = ( z15, y15, x15, z14, y14, x14, z13, y13, x13, z12, y12, x12, z11, y11, x11, z10 ) const __m128i vxyz2 = _mm_castps_si128( _mm_shuffle_ps(_mm_castsi128_ps(vtemp0), _mm_castsi128_ps(vtemp2), _MM_SHUFFLE(3, 1, 3, 1))); o = (uint8_t*) ((uintptr_t) o + address_increment * 3); _mm_storeu_si128((__m128i*) o, vxyz0); _mm_storeu_si128((__m128i*) o + 1, vxyz1); _mm_storeu_si128((__m128i*) o + 2, vxyz2); } } else { do { const uint8_t vx = *x++; const uint8_t vy = *y++; const uint8_t vz = *z++; o[0] = vx; o[1] = vy; o[2] = vz; o += 3; } while (--n != 0); } }
8,127
57.898551
126
c
XNNPACK
XNNPACK-master/src/x8-zip/x8-zip-x4-neon.c
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <xnnpack/zip.h> void xnn_x8_zip_x4_ukernel__neon( size_t n, const uint8_t* input, uint8_t* output) { const uint8_t* x = input; const uint8_t* y = (const uint8_t*) ((uintptr_t) x + n); const uint8_t* z = (const uint8_t*) ((uintptr_t) y + n); const uint8_t* w = (const uint8_t*) ((uintptr_t) z + n); uint8_t* o = output; if (n >= 8) { do { uint8x8x4_t vxyzw; vxyzw.val[0] = vld1_u8(x); x += 8; vxyzw.val[1] = vld1_u8(y); y += 8; vxyzw.val[2] = vld1_u8(z); z += 8; vxyzw.val[3] = vld1_u8(w); w += 8; vst4_u8(o, vxyzw); o += 32; n -= 8; } while (n >= 8); if (n != 0) { const size_t address_increment = n - 8; uint8x8x4_t vxyzw; vxyzw.val[0] = vld1_u8((const uint8_t*) ((uintptr_t) x + address_increment)); vxyzw.val[1] = vld1_u8((const uint8_t*) ((uintptr_t) y + address_increment)); vxyzw.val[2] = vld1_u8((const uint8_t*) ((uintptr_t) z + address_increment)); vxyzw.val[3] = vld1_u8((const uint8_t*) ((uintptr_t) w + address_increment)); vst4_u8((uint8_t*) ((uintptr_t) o + address_increment * 4), vxyzw); } } else { do { const uint8_t vx = *x++; const uint8_t vy = *y++; const uint8_t vz = *z++; const uint8_t vw = *w++; o[0] = vx; o[1] = vy; o[2] = vz; o[3] = vw; o += 4; } while (--n != 0); } }
1,665
27.724138
83
c
XNNPACK
XNNPACK-master/src/x8-zip/x8-zip-x4-scalar.c
// Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <assert.h> #include <xnnpack/zip.h> void xnn_x8_zip_x4_ukernel__scalar( size_t n, const uint8_t* input, uint8_t* output) { assert(n != 0); const uint8_t* x = input; const uint8_t* y = (const uint8_t*) ((uintptr_t) x + n); const uint8_t* z = (const uint8_t*) ((uintptr_t) y + n); const uint8_t* w = (const uint8_t*) ((uintptr_t) z + n); uint8_t* o = output; do { const uint8_t vx = *x++; const uint8_t vy = *y++; const uint8_t vz = *z++; const uint8_t vw = *w++; o[0] = vx; o[1] = vy; o[2] = vz; o[3] = vw; o += 4; n -= sizeof(uint8_t); } while (n != 0); }
804
20.184211
72
c
XNNPACK
XNNPACK-master/src/x8-zip/x8-zip-x4-sse2.c
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <emmintrin.h> #include <xnnpack/zip.h> void xnn_x8_zip_x4_ukernel__sse2( size_t n, const uint8_t* input, uint8_t* output) { const uint8_t* x = input; const uint8_t* y = (const uint8_t*) ((uintptr_t) x + n); const uint8_t* z = (const uint8_t*) ((uintptr_t) y + n); const uint8_t* w = (const uint8_t*) ((uintptr_t) z + n); uint8_t* o = output; if (n >= 16) { do { const __m128i vx = _mm_loadu_si128((const __m128i*) x); x += 16; const __m128i vy = _mm_loadu_si128((const __m128i*) y); y += 16; const __m128i vz = _mm_loadu_si128((const __m128i*) z); z += 16; const __m128i vw = _mm_loadu_si128((const __m128i*) w); w += 16; const __m128i vxy_lo = _mm_unpacklo_epi8(vx, vy); const __m128i vxy_hi = _mm_unpackhi_epi8(vx, vy); const __m128i vzw_lo = _mm_unpacklo_epi8(vz, vw); const __m128i vzw_hi = _mm_unpackhi_epi8(vz, vw); const __m128i vxyzw0 = _mm_unpacklo_epi16(vxy_lo, vzw_lo); const __m128i vxyzw1 = _mm_unpackhi_epi16(vxy_lo, vzw_lo); const __m128i vxyzw2 = _mm_unpacklo_epi16(vxy_hi, vzw_hi); const __m128i vxyzw3 = _mm_unpackhi_epi16(vxy_hi, vzw_hi); _mm_storeu_si128((__m128i*) o, vxyzw0); _mm_storeu_si128((__m128i*) o + 1, vxyzw1); _mm_storeu_si128((__m128i*) o + 2, vxyzw2); _mm_storeu_si128((__m128i*) o + 3, vxyzw3); o = (void*) ((uintptr_t) o + 64); n -= 16; } while (n >= 16); if (n != 0) { const size_t address_increment = n - 16; const __m128i vx = _mm_loadu_si128((const __m128i*) ((uintptr_t) x + address_increment)); const __m128i vy = _mm_loadu_si128((const __m128i*) ((uintptr_t) y + address_increment)); const __m128i vz = _mm_loadu_si128((const __m128i*) ((uintptr_t) z + address_increment)); const __m128i vw = _mm_loadu_si128((const __m128i*) ((uintptr_t) w + address_increment)); const __m128i vxy_lo = _mm_unpacklo_epi8(vx, vy); const __m128i vxy_hi = _mm_unpackhi_epi8(vx, vy); const __m128i vzw_lo = _mm_unpacklo_epi8(vz, vw); const __m128i vzw_hi = _mm_unpackhi_epi8(vz, vw); const __m128i vxyzw0 = _mm_unpacklo_epi16(vxy_lo, vzw_lo); const __m128i vxyzw1 = _mm_unpackhi_epi16(vxy_lo, vzw_lo); const __m128i vxyzw2 = _mm_unpacklo_epi16(vxy_hi, vzw_hi); const __m128i vxyzw3 = _mm_unpackhi_epi16(vxy_hi, vzw_hi); o = (void*) ((uintptr_t) o + address_increment * 4); _mm_storeu_si128((__m128i*) o, vxyzw0); _mm_storeu_si128((__m128i*) o + 1, vxyzw1); _mm_storeu_si128((__m128i*) o + 2, vxyzw2); _mm_storeu_si128((__m128i*) o + 3, vxyzw3); } } else { do { const uint8_t vx = *x++; const uint8_t vy = *y++; const uint8_t vz = *z++; const uint8_t vw = *w++; o[0] = vx; o[1] = vy; o[2] = vz; o[3] = vw; o += 4; } while (--n != 0); } }
3,163
36.666667
95
c
XNNPACK
XNNPACK-master/src/x8-zip/x8-zip-xm-neon.c
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <arm_neon.h> #include <xnnpack/zip.h> void xnn_x8_zip_xm_ukernel__neon( size_t n, size_t m, const uint8_t* input, uint8_t* output) { const uint8_t* w = input; const size_t input_increment = n * 3; const size_t output_increment = 4 - m * n; const uint8_t* last_input = w + n * (m - 1); uint8_t* last_output = (uint8_t*) ((uintptr_t) output + (m - 4)); if (n >= 8) { for (size_t i = 0; i < m; i += 4) { size_t k = n; w = (const uint8_t*) ((uintptr_t) w + input_increment); if (w >= last_input) { w = last_input; } const uint8_t* z = (const uint8_t*) ((uintptr_t) w - n); const uint8_t* y = (const uint8_t*) ((uintptr_t) z - n); const uint8_t* x = (const uint8_t*) ((uintptr_t) y - n); while (k >= 8) { const uint8x8_t vx = vld1_u8(x); x += 8; const uint8x8_t vy = vld1_u8(y); y += 8; const uint8x8_t vz = vld1_u8(z); z += 8; const uint8x8_t vw = vld1_u8(w); w += 8; const uint8x8x2_t vxy = vzip_u8(vx, vy); const uint8x8x2_t vzw = vzip_u8(vz, vw); const uint16x4x2_t vxyzw_lo = vzip_u16(vreinterpret_u16_u8(vxy.val[0]), vreinterpret_u16_u8(vzw.val[0])); const uint16x4x2_t vxyzw_hi = vzip_u16(vreinterpret_u16_u8(vxy.val[1]), vreinterpret_u16_u8(vzw.val[1])); vst1_lane_u32((void*) output, vreinterpret_u32_u16(vxyzw_lo.val[0]), 0); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vreinterpret_u32_u16(vxyzw_lo.val[0]), 1); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vreinterpret_u32_u16(vxyzw_lo.val[1]), 0); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vreinterpret_u32_u16(vxyzw_lo.val[1]), 1); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vreinterpret_u32_u16(vxyzw_hi.val[0]), 0); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vreinterpret_u32_u16(vxyzw_hi.val[0]), 1); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vreinterpret_u32_u16(vxyzw_hi.val[1]), 0); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vreinterpret_u32_u16(vxyzw_hi.val[1]), 1); output = (uint8_t*) ((uintptr_t) output + m); k -= 8; } if (k != 0) { const size_t address_increment = k - 8; x = (const uint8_t*) ((uintptr_t) x + address_increment); y = (const uint8_t*) ((uintptr_t) y + address_increment); z = (const uint8_t*) ((uintptr_t) z + address_increment); w = (const uint8_t*) ((uintptr_t) w + address_increment); const int64x1_t vshift = vmov_n_s64(8 * address_increment); const uint64x1_t vx = vshl_u64(vreinterpret_u64_u8(vld1_u8(x)), vshift); const uint64x1_t vy = vshl_u64(vreinterpret_u64_u8(vld1_u8(y)), vshift); const uint64x1_t vz = vshl_u64(vreinterpret_u64_u8(vld1_u8(z)), vshift); const uint64x1_t vw = vshl_u64(vreinterpret_u64_u8(vld1_u8(w)), vshift); w += 8; const uint8x8x2_t vxy = vzip_u8(vreinterpret_u8_u64(vx), vreinterpret_u8_u64(vy)); const uint8x8x2_t vzw = vzip_u8(vreinterpret_u8_u64(vz), vreinterpret_u8_u64(vw)); const uint16x4x2_t vxyzw_lo = vzip_u16(vreinterpret_u16_u8(vxy.val[0]), vreinterpret_u16_u8(vzw.val[0])); const uint16x4x2_t vxyzw_hi = vzip_u16(vreinterpret_u16_u8(vxy.val[1]), vreinterpret_u16_u8(vzw.val[1])); uint32x2_t vxyzw0 = vreinterpret_u32_u16(vxyzw_lo.val[0]); uint32x2_t vxyzw1 = vreinterpret_u32_u16(vxyzw_lo.val[1]); uint32x2_t vxyzw2 = vreinterpret_u32_u16(vxyzw_hi.val[0]); uint32x2_t vxyzw3 = vreinterpret_u32_u16(vxyzw_hi.val[1]); if (k & 4) { vst1_lane_u32((void*) output, vxyzw0, 0); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vxyzw0, 1); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vxyzw1, 0); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vxyzw1, 1); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = vxyzw2; vxyzw1 = vxyzw3; } if (k & 2) { vst1_lane_u32((void*) output, vxyzw0, 0); output = (uint8_t*) ((uintptr_t) output + m); vst1_lane_u32((void*) output, vxyzw0, 1); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = vxyzw1; } if (k & 1) { vst1_lane_u32((void*) output, vxyzw0, 0); output = (uint8_t*) ((uintptr_t) output + m); } } output = (uint8_t*) ((uintptr_t) output + output_increment); if (output > last_output) { output = last_output; } } } else { const uint8_t* i = input; uint8_t* o = output; size_t k = n; do { size_t l = m; const uint8_t* ii = i++; do { *o++ = *ii; ii += n; } while (--l != 0); } while (--k != 0); } }
5,432
36.468966
113
c
XNNPACK
XNNPACK-master/src/x8-zip/x8-zip-xm-sse2.c
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #include <emmintrin.h> #include <xnnpack/zip.h> #include <xnnpack/unaligned.h> void xnn_x8_zip_xm_ukernel__sse2( size_t n, size_t m, const uint8_t* input, uint8_t* output) { const uint8_t* w = input; const size_t input_increment = n * 3; const size_t output_increment = 4 - m * n; const uint8_t* last_input = w + n * (m - 1); uint8_t* last_output = (uint8_t*) ((uintptr_t) output + (m - 4)); if (n >= 8) { for (size_t i = 0; i < m; i += 4) { size_t k = n; w = (const uint8_t*) ((uintptr_t) w + input_increment); if (w >= last_input) { w = last_input; } const uint8_t* z = (const uint8_t*) ((uintptr_t) w - n); const uint8_t* y = (const uint8_t*) ((uintptr_t) z - n); const uint8_t* x = (const uint8_t*) ((uintptr_t) y - n); while (k >= 16) { const __m128i vx = _mm_loadu_si128((const __m128i*) x); x += 16; const __m128i vy = _mm_loadu_si128((const __m128i*) y); y += 16; const __m128i vz = _mm_loadu_si128((const __m128i*) z); z += 16; const __m128i vw = _mm_loadu_si128((const __m128i*) w); w += 16; const __m128i vxy_lo = _mm_unpacklo_epi8(vx, vy); const __m128i vxy_hi = _mm_unpackhi_epi8(vx, vy); const __m128i vzw_lo = _mm_unpacklo_epi8(vz, vw); const __m128i vzw_hi = _mm_unpackhi_epi8(vz, vw); __m128i vxyzw0 = _mm_unpacklo_epi16(vxy_lo, vzw_lo); __m128i vxyzw1 = _mm_unpackhi_epi16(vxy_lo, vzw_lo); __m128i vxyzw2 = _mm_unpacklo_epi16(vxy_hi, vzw_hi); __m128i vxyzw3 = _mm_unpackhi_epi16(vxy_hi, vzw_hi); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_unpackhi_epi64(vxyzw0, vxyzw0); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw1)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw1 = _mm_shufflelo_epi16(vxyzw1, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw1)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw1 = _mm_unpackhi_epi64(vxyzw1, vxyzw1); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw1)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw1 = _mm_shufflelo_epi16(vxyzw1, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw1)); output = (uint8_t*) ((uintptr_t) output + m); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw2)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw2 = _mm_shufflelo_epi16(vxyzw2, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw2)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw2 = _mm_unpackhi_epi64(vxyzw2, vxyzw2); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw2)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw2 = _mm_shufflelo_epi16(vxyzw2, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw2)); output = (uint8_t*) ((uintptr_t) output + m); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw3)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw3 = _mm_shufflelo_epi16(vxyzw3, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw3)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw3 = _mm_unpackhi_epi64(vxyzw3, vxyzw3); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw3)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw3 = _mm_shufflelo_epi16(vxyzw3, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw3)); output = (uint8_t*) ((uintptr_t) output + m); k -= 16; }; if (k >= 8) { const __m128i vx = _mm_loadl_epi64((const __m128i*) x); x += 8; const __m128i vy = _mm_loadl_epi64((const __m128i*) y); y += 8; const __m128i vz = _mm_loadl_epi64((const __m128i*) z); z += 8; const __m128i vw = _mm_loadl_epi64((const __m128i*) w); w += 8; const __m128i vxy = _mm_unpacklo_epi8(vx, vy); const __m128i vzw = _mm_unpacklo_epi8(vz, vw); __m128i vxyzw0 = _mm_unpacklo_epi16(vxy, vzw); __m128i vxyzw1 = _mm_unpackhi_epi16(vxy, vzw); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_unpackhi_epi64(vxyzw0, vxyzw0); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw1)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw1 = _mm_shufflelo_epi16(vxyzw1, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw1)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw1 = _mm_unpackhi_epi64(vxyzw1, vxyzw1); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw1)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw1 = _mm_shufflelo_epi16(vxyzw1, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw1)); output = (uint8_t*) ((uintptr_t) output + m); k -= 8; } if (k != 0) { const size_t address_decrement = 8 - k; x -= address_decrement; y -= address_decrement; z -= address_decrement; w -= address_decrement; const __m128i vshift = _mm_cvtsi32_si128((int) address_decrement * 8); const __m128i vx = _mm_srl_epi64(_mm_loadl_epi64((const __m128i*) x), vshift); const __m128i vy = _mm_srl_epi64(_mm_loadl_epi64((const __m128i*) y), vshift); const __m128i vz = _mm_srl_epi64(_mm_loadl_epi64((const __m128i*) z), vshift); const __m128i vw = _mm_srl_epi64(_mm_loadl_epi64((const __m128i*) w), vshift); w += 8; const __m128i vxy = _mm_unpacklo_epi8(vx, vy); const __m128i vzw = _mm_unpacklo_epi8(vz, vw); __m128i vxyzw0 = _mm_unpacklo_epi16(vxy, vzw); __m128i vxyzw1 = _mm_unpackhi_epi16(vxy, vzw); if (k & 4) { unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_unpackhi_epi64(vxyzw0, vxyzw0); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = vxyzw1; } if (k & 2) { unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_shufflelo_epi16(vxyzw0, _MM_SHUFFLE(3, 2, 3, 2)); unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); vxyzw0 = _mm_unpackhi_epi64(vxyzw0, vxyzw0); } if (k & 1) { unaligned_store_u32(output, (uint32_t) _mm_cvtsi128_si32(vxyzw0)); output = (uint8_t*) ((uintptr_t) output + m); } } output = (uint8_t*) ((uintptr_t) output + output_increment); if (output > last_output) { output = last_output; } } } else { const uint8_t* i = input; uint8_t* o = output; size_t k = n; do { size_t l = m; const uint8_t* ii = i++; do { *o++ = *ii; ii += n; } while (--l != 0); } while (--k != 0); } }
9,403
44.211538
86
c
XNNPACK
XNNPACK-master/src/xnnpack/aarch32-assembler.h
// Copyright 2021 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <cstddef> #include <cstdint> #include <initializer_list> #include <xnnpack/assembler.h> // MSVC defines these tokens using macros, causing name collisions. We use these name in our assembler, so undef them. // These macros are pulled in via xnnpack/math.h -> intrin.h -> arm_neon.h. #include <xnnpack/math.h> #ifdef vabs_f32 #undef vabs_f32 #endif #ifdef vadd_f32 #undef vadd_f32 #endif #ifdef vcvt_f32_s32 #undef vcvt_f32_s32 #endif #ifdef vcvt_s32_f32 #undef vcvt_s32_f32 #endif #ifdef vcvtn_s32_f32 #undef vcvtn_s32_f32 #endif #ifdef vmax_f32 #undef vmax_f32 #endif #ifdef vmax_s8 #undef vmax_s8 #endif #ifdef vmin_f32 #undef vmin_f32 #endif #ifdef vmin_s8 #undef vmin_s8 #endif #ifdef vmla_f32 #undef vmla_f32 #endif #ifdef vmlal_s16 #undef vmlal_s16 #endif #ifdef vmovl_s8 #undef vmovl_s8 #endif #ifdef vmul_f32 #undef vmul_f32 #endif #ifdef vneg_f32 #undef vneg_f32 #endif #ifdef vqadd_s16 #undef vqadd_s16 #endif #ifdef vqdmulh_s32 #undef vqdmulh_s32 #endif #ifdef vqmovn_s16 #undef vqmovn_s16 #endif #ifdef vqmovn_s32 #undef vqmovn_s32 #endif #ifdef vqshl_s32 #undef vqshl_s32 #endif #ifdef vrshl_s32 #undef vrshl_s32 #endif namespace xnnpack { namespace aarch32 { // Special values used to check that callee-saved registers are properly saved. // Low 8 bits should be 0 to encode register code. constexpr uint32_t kRRegisterCorruptValue = UINT32_C(0xDEADBE00); constexpr uint32_t kSRegisterCorruptValue = UINT32_C(0x7FF00000); constexpr uint8_t kRegisterCorruptMask = UINT8_C(0xFF); // Instruction used to align code, is a nop. constexpr uint32_t kAlignInstruction = 0xE320F000; enum class SpecialFPRegister { kFPSCR = 1, }; constexpr SpecialFPRegister FPSCR = SpecialFPRegister::kFPSCR; struct CoreRegister { uint8_t code; }; constexpr CoreRegister r0{0}; constexpr CoreRegister r1{1}; constexpr CoreRegister r2{2}; constexpr CoreRegister r3{3}; constexpr CoreRegister r4{4}; constexpr CoreRegister r5{5}; constexpr CoreRegister r6{6}; constexpr CoreRegister r7{7}; constexpr CoreRegister r8{8}; constexpr CoreRegister r9{9}; constexpr CoreRegister r10{10}; constexpr CoreRegister r11{11}; constexpr CoreRegister r12{12}; constexpr CoreRegister r13{13}; constexpr CoreRegister r14{14}; constexpr CoreRegister r15{15}; constexpr CoreRegister sp = r13; constexpr CoreRegister lr = r14; constexpr CoreRegister pc = r15; constexpr CoreRegister APSR_nzcv = r15; static inline bool operator==(const CoreRegister lhs, const CoreRegister rhs) { return lhs.code == rhs.code; } struct CoreRegisterList { CoreRegisterList(std::initializer_list<CoreRegister> rs) { for (auto r : rs) { list |= 1 << r.code; } } bool has_more_than_one_register() { return (list & (list - 1)) != 0; } // Bit i is set if CoreRegister is in the list. uint16_t list = 0; }; static inline bool operator==(int i, CoreRegisterList registers) { return i == registers.list; } struct SRegister { uint8_t code; uint8_t d() const { return code & 0x1; } uint8_t vd() const { return (code & 0x1e) >> 1; } }; static inline bool operator==(const SRegister lhs, const SRegister rhs) { return lhs.code == rhs.code; } constexpr SRegister s0{0}; constexpr SRegister s1{1}; constexpr SRegister s2{2}; constexpr SRegister s3{3}; constexpr SRegister s4{4}; constexpr SRegister s5{5}; constexpr SRegister s6{6}; constexpr SRegister s7{7}; constexpr SRegister s8{8}; constexpr SRegister s9{9}; constexpr SRegister s10{10}; constexpr SRegister s11{11}; constexpr SRegister s12{12}; constexpr SRegister s13{13}; constexpr SRegister s14{14}; constexpr SRegister s15{15}; constexpr SRegister s16{16}; constexpr SRegister s17{17}; constexpr SRegister s18{18}; constexpr SRegister s19{19}; constexpr SRegister s20{20}; constexpr SRegister s21{21}; constexpr SRegister s22{22}; constexpr SRegister s23{23}; constexpr SRegister s24{24}; constexpr SRegister s25{25}; constexpr SRegister s26{26}; constexpr SRegister s27{27}; constexpr SRegister s28{28}; constexpr SRegister s29{29}; constexpr SRegister s30{30}; constexpr SRegister s31{31}; // Define DRegisterLane before DRegister so that we can have the operator[] overloading for nice syntax. struct DRegisterLane { uint8_t code; uint8_t lane; uint8_t d() const { return (code & 0x10) >> 4; } uint8_t vd() const { return code & 0xf; } }; static inline bool operator==(const DRegisterLane lhs, const DRegisterLane rhs) { return lhs.code == rhs.code && lhs.lane == rhs.lane; } struct DRegister { uint8_t code; uint8_t d() const { return (code & 0x10) >> 4; } uint8_t vd() const { return code & 0xf; } SRegister low() const { return SRegister{uint8_t(code * 2)}; } SRegister high() const { return SRegister{uint8_t(code * 2 + 1)}; } DRegisterLane operator[](std::size_t pos) const { return DRegisterLane{code, static_cast<uint8_t>(pos)}; } }; static inline bool operator==(const DRegister lhs, const DRegister rhs) { return lhs.code == rhs.code; } constexpr DRegister d0{0}; constexpr DRegister d1{1}; constexpr DRegister d2{2}; constexpr DRegister d3{3}; constexpr DRegister d4{4}; constexpr DRegister d5{5}; constexpr DRegister d6{6}; constexpr DRegister d7{7}; constexpr DRegister d8{8}; constexpr DRegister d9{9}; constexpr DRegister d10{10}; constexpr DRegister d11{11}; constexpr DRegister d12{12}; constexpr DRegister d13{13}; constexpr DRegister d14{14}; constexpr DRegister d15{15}; constexpr DRegister d16{16}; constexpr DRegister d17{17}; constexpr DRegister d18{18}; constexpr DRegister d19{19}; constexpr DRegister d20{20}; constexpr DRegister d21{21}; constexpr DRegister d22{22}; constexpr DRegister d23{23}; constexpr DRegister d24{24}; constexpr DRegister d25{25}; constexpr DRegister d26{26}; constexpr DRegister d27{27}; constexpr DRegister d28{28}; constexpr DRegister d29{29}; constexpr DRegister d30{30}; constexpr DRegister d31{31}; struct QRegister { uint8_t code; // Encode code * 2. uint8_t d() const { return (code & 0x8) >> 3; } uint8_t vd() const { return (code & 0x7) << 1; } DRegister low() const { return DRegister{uint8_t(code * 2)}; } DRegister high() const { return DRegister{uint8_t(code * 2 + 1)}; } }; static inline bool operator==(const QRegister lhs, const QRegister rhs) { return lhs.code == rhs.code; } constexpr QRegister q0{0}; constexpr QRegister q1{1}; constexpr QRegister q2{2}; constexpr QRegister q3{3}; constexpr QRegister q4{4}; constexpr QRegister q5{5}; constexpr QRegister q6{6}; constexpr QRegister q7{7}; constexpr QRegister q8{8}; constexpr QRegister q9{9}; constexpr QRegister q10{10}; constexpr QRegister q11{11}; constexpr QRegister q12{12}; constexpr QRegister q13{13}; constexpr QRegister q14{14}; constexpr QRegister q15{15}; // SIMD register lists are used in a more restrictive way, compared to core // registers, only consecutive registers are used as an operand to instruction. template <typename RegType> struct ConsecutiveRegisterList { // End must be >= start. ConsecutiveRegisterList(RegType s, RegType end) : start(s), length(end.code - s.code + 1) {} explicit ConsecutiveRegisterList(RegType s, int len) : start(s), length(len) {} // NOLINTNEXTLINE(google-explicit-constructor) ConsecutiveRegisterList(RegType start) : ConsecutiveRegisterList(start, start) {} RegType start; uint8_t length; }; // Specific struct for VLD2 and VLD3 register list operand. struct VLoadStoreRegList { VLoadStoreRegList(DRegister reg1, DRegister reg2) : reg1(reg1), reg2(reg2) { if (reg1.code == reg2.code - 2) { double_spaced = true; } else { double_spaced = false; } } VLoadStoreRegList(DRegister reg1, DRegister reg2, DRegister reg3) : reg1(reg1), reg2(reg2), reg3(reg3) { if (reg1.code == reg2.code - 2) { double_spaced = true; } else { double_spaced = false; } } DRegister reg1; DRegister reg2; DRegister reg3; bool double_spaced; }; using SRegisterList = ConsecutiveRegisterList<SRegister>; using DRegisterList = ConsecutiveRegisterList<DRegister>; static inline SRegisterList operator-(const SRegister lhs, const SRegister rhs) { return SRegisterList(lhs, rhs); } static inline DRegisterList operator-(const DRegister lhs, const DRegister rhs) { return DRegisterList(lhs, rhs); } struct QRegisterList { // NOLINTNEXTLINE(google-explicit-constructor) QRegisterList(QRegister s) : start(s), length(1) {} QRegisterList(QRegister s, QRegister end) : start(s), length(end.code - s.code + 1) {} // Explicit conversion to DRegisterList. explicit operator DRegisterList() const { return DRegisterList({static_cast<uint8_t>(start.code * 2)}, length * 2); } QRegister start; uint8_t length; }; static inline QRegisterList operator-(const QRegister lhs, const QRegister rhs) { return QRegisterList(lhs, rhs); } // A8.5 Addressing modes for memory access. enum class AddressingMode { // [<Rn>, <offset>], offset applied to address in Rn. kOffset, // Pre-indexed not used, so not implemented. // [<Rn>], <offset>, address from Rn, offset applied, written back to Rn. kPostIndexed, }; // Memory operands, operands for memory access instructions. See // "MemOperandHelper mem" for a nicer syntax that is closer to assembly. class MemOperand { public: MemOperand(CoreRegister rn, int32_t offset) : mode_(AddressingMode::kOffset), rn_(rn), offset_(offset) {} MemOperand(CoreRegister rn, int32_t offset, AddressingMode mode) : mode_(mode), rn_(rn), offset_(offset) {} CoreRegister base() const { return rn_; } int32_t offset() const { return offset_; } AddressingMode mode() const { return mode_; } // These are bits used for encoding, named based on the encoding description. int32_t u() { return static_cast<int32_t>(offset_ >= 0); } int32_t p() { return static_cast<int32_t>(mode_ != AddressingMode::kPostIndexed); } // Note, kPostIndexed will write back, but doesn't need to set bit w. int32_t w() { return 0; } // Overload postfix increment to indicate a post-indexed addressing mode for load/stores. MemOperand operator++(int) { mode_ = AddressingMode::kPostIndexed; return *this; } private: AddressingMode mode_; CoreRegister rn_; int32_t offset_; }; static inline bool operator==(const MemOperand lhs, const MemOperand rhs) { return lhs.mode() == rhs.mode() && lhs.base() == rhs.base() && lhs.offset() == rhs.offset(); } static inline MemOperand operator,(CoreRegister r, int32_t offset) { return MemOperand(r, offset); } // Helper struct for some syntax sugar to look like native assembly, see mem. struct MemOperandHelper { MemOperand operator[](MemOperand op) const { return op; } MemOperand operator[](CoreRegister r) const { return MemOperand(r, 0); } }; // Use "mem" (and its overload of array subscript operator) to get some syntax // that looks closer to native assembly when accessing memory. For example: // - ldr(r0, mem[rn, offset]); // offset // - ldr(r0, mem[rn], offset); // post-indexed constexpr MemOperandHelper mem; // Conditional execution, only support AL (always) for now. enum Condition : uint32_t { kEQ = 0x00000000, kNE = 0x10000000, kCS = 0x20000000, kCC = 0x30000000, kMI = 0x40000000, kPL = 0x50000000, kVS = 0x60000000, kVC = 0x70000000, kHI = 0x80000000, kLS = 0x90000000, kGE = 0xa0000000, kLT = 0xB0000000, kGT = 0xC0000000, kLE = 0xD0000000, kAL = 0xE0000000, kHS = kCS, kLO = kCC, }; enum DataSize { k8 = 0, k16 = 1, k32 = 2, }; // A simple AAarch32 assembler. class Assembler : public AssemblerBase { public: using AssemblerBase::AssemblerBase; void add(CoreRegister rn, CoreRegister rm) { add(rn, rn, rm); } void add(CoreRegister rd, CoreRegister rn, CoreRegister rm); // Only support uint8_t immediates for now, it simplifies encoding. void add(CoreRegister rd, CoreRegister rn, uint8_t imm); void adds(CoreRegister rd, CoreRegister rn, uint8_t imm); void and_(CoreRegister rd, CoreRegister rn, uint8_t imm); void b(Label& l) { b(kAL, l); } void beq(Label& l) { b(kEQ, l); } void bne(Label& l) { b(kNE, l); } void bhi(Label& l) { b(kHI, l); } void bhs(Label& l) { b(kHS, l); } void blo(Label& l) { b(kLO, l); } void blx(CoreRegister rm); void bic(CoreRegister rd, CoreRegister rn, uint8_t imm); void bx(CoreRegister rm); // Cmp supports a subset of uint32_t offsets, see "A5.2.4 Modified immediate // constants in ARM instructions", for simplicity we start with uint8_t, which // is fully representation using a "rotation" of 0. void cmp(CoreRegister rn, uint8_t imm); void cmp(CoreRegister rn, CoreRegister rm); void ldr(CoreRegister rt, MemOperand operand, int32_t offset); void ldr(CoreRegister rt, MemOperand operand); // LDRD <Rt>, <Rt2>, [<Rn>{, #+/-<imm>}]. void ldrd(CoreRegister rt, CoreRegister rt2, MemOperand op); void mov(CoreRegister rd, CoreRegister rm); void mov(CoreRegister rd, uint16_t imm); void movt(CoreRegister rd, uint16_t imm); void moveq(CoreRegister rd, CoreRegister rm) { mov(kEQ, rd, rm); } void movlo(CoreRegister rd, CoreRegister rm) { mov(kLO, rd, rm); } void movls(CoreRegister rd, CoreRegister rm) { mov(kLS, rd, rm); } void nop(); void pld(MemOperand operand); void pop(CoreRegisterList regs); void push(CoreRegisterList regs); void str(CoreRegister rt, MemOperand op); void sub(CoreRegister rd, CoreRegister rn, uint8_t imm); void sub(CoreRegister rd, CoreRegister rn, CoreRegister rm); // Only support uint8_t immediates for now, it simplifies encoding. void subs(CoreRegister rd, CoreRegister rn, uint8_t imm); void tst(CoreRegister rn, uint8_t imm); // SIMD instructions. void vabs_f32(QRegister qd, QRegister qm); void vadd_f32(QRegister qd, QRegister qn, QRegister qm); void vcmpe_f32(SRegister sd, SRegister sm); void vcvt_f32_s32(QRegister qd, QRegister qm); void vcvt_s32_f32(QRegister qd, QRegister qm); void vcvtn_s32_f32(QRegister qd, QRegister qm); void vdup_8(QRegister qd, DRegisterLane dm) { vdup(k8, qd, dm); } void vdup_16(QRegister qd, DRegisterLane dm) { vdup(k16, qd, dm); } void vdup_32(QRegister qd, DRegisterLane dm) { vdup(k32, qd, dm); } void vext_8(QRegister qd, QRegister qn, QRegister qm, uint8_t imm4); // VLD1.8 <list>, [<Rn>]{!} (multiple single elements). void vld1_8(DRegisterList regs, MemOperand op) { vld1(k8, regs, op); } void vld1_8(DRegisterList regs, MemOperand op, CoreRegister rm) { vld1(k8, regs, op, rm); } void vld1_8(QRegisterList regs, MemOperand op) { vld1(k8, static_cast<DRegisterList>(regs), op); } // VLD1.32 <list>, [<Rn>]{!} (multiple single elements). void vld1_32(DRegisterList regs, MemOperand op) { vld1(k32, regs, op); } void vld1_32(QRegisterList regs, MemOperand op) { vld1(k32, static_cast<DRegisterList>(regs), op); } // VLD1.32 <list>, [<Rn>]{!} (single element to one lane). void vld1_32(DRegisterLane dd, MemOperand op); // VLD1.32 <list>, [<Rn>]{!} (single element to all lanes). // We cannot differentiate the register list in C++ syntax, so use an instruction name similar to AArch64 LD1R. void vld1r_32(DRegisterList regs, MemOperand op); void vld2r_32(VLoadStoreRegList regs, MemOperand op); void vld3r_32(VLoadStoreRegList regs, MemOperand op); // VLDM <Rn>{!}, <list> (IA). void vldm(MemOperand rn, SRegisterList regs); void vldm(MemOperand rn, DRegisterList regs); void vldr(SRegister sd, MemOperand op); void vldr(DRegister dd, MemOperand op); void vmax_f32(QRegister qd, QRegister qn, QRegister qm); void vmax_s8(QRegister qd, QRegister qn, QRegister qm); void vmin_f32(QRegister qd, QRegister qn, QRegister qm); void vmin_s8(QRegister qd, QRegister qn, QRegister qm); // VMLA.F32 <Sd>, <Sn>, <Sm> void vmla_f32(SRegister sd, SRegister sn, SRegister sm); // VMLA.F32 <Qd>, <Qn>, <Dm[x]> void vmla_f32(QRegister qd, QRegister qn, DRegisterLane dm); // VMLAL.S16 <Qd>, <Dn>, <Dm[x]> void vmlal_s16(QRegister qd, DRegister dn, DRegisterLane dm); // VMOV.I32 <Qd>, #<imm>; encoding A1 void vmov_i32(QRegister qd, uint8_t imm); // VMOV.F32 <Qd>, #<imm>; encoding A1 void vmov(QRegister qd, uint8_t imm); // VMOV <Rt>, <Sn>; encoding A1. void vmov(CoreRegister rt, SRegister sn); // VMOV <Sn>, <Rt>; encoding A1. void vmov(SRegister sn, CoreRegister rt); // VMOV.F32 <Sd>, <Sm>; encoding A2. void vmov(SRegister sd, SRegister sm); // VMOV <Dm>, <Rt>, <Rt2>; encoding A1. void vmov(DRegister dm, CoreRegister rt, CoreRegister rt2); // VMOV <Rt>, <Rt2>, <Dm>; encoding A1. void vmov(CoreRegister rt, CoreRegister rt2, DRegister dm); // VMOV <Dd>, <Dm>; encoding A1. void vmov(DRegister dd, DRegister dm); // VMOV <Qd>, <Qm>; encoding A1. void vmov(QRegister qd, QRegister qm); // VMOV_F32 <Sd>, <Sm> void vmov_f32(SRegister sd, SRegister sm) { vmov_f32(kAL, sd, sm); } void vmovpl_f32(SRegister sd, SRegister sm) { vmov_f32(kPL, sd, sm); } void vmovmi_f32(SRegister sd, SRegister sm) { vmov_f32(kMI, sd, sm); } // VMOV_F64 <Dd>, <Dm> void vmov_f64(DRegister dd, DRegister dm); // VMOVL.S8 <Qd>, <Dm> void vmovl_s8(QRegister qd, DRegister dm); void vmrs(CoreRegister rt, SpecialFPRegister spec_reg); void vmul_f32(QRegister qd, QRegister qn, QRegister qm); // VMUL.F32 <Qd>, <Qn>, <Dm[x]> void vmul_f32(QRegister qd, QRegister qn, DRegisterLane dm); void vneg_f32(QRegister qd, QRegister qm); void vpop(DRegisterList regs); void vpush(DRegisterList regs); void vpush(SRegisterList regs); void vqadd_s16(QRegister qd, QRegister qn, QRegister qm); void vqdmulh_s32(QRegister qd, QRegister qn, DRegisterLane dm); void vqmovn_s16(DRegister dd, QRegister qm); void vqmovn_s32(DRegister dd, QRegister qm); void vqshl_s32(QRegister qd, QRegister qm, QRegister qn); void vrshl_s32(QRegister qd, QRegister qm, QRegister qn); void vsdot_s8(QRegister qd, QRegister qn, DRegisterLane dm); // VST1.8 <list>, [<Rn>]{!} (multiple single elements). void vst1_8(DRegisterList regs, MemOperand op) { vst1(k8, regs, op); } // VST1.8 <list>, [<Rn>]{!}, <Rm> (multiple single elements). void vst1_8(DRegisterList regs, MemOperand op, CoreRegister rm) { vst1(k8, regs, op, rm); } // VST1.8 <list>, [<Rn>]{!} (single element form one lane). void vst1_8(DRegisterLane dd, MemOperand op) { vst1(k8, dd, op); } // VST1.16 <list>, [<Rn>]{!} (multiple single elements). void vst1_16(DRegisterList regs, MemOperand op) { vst1(k16, regs, op); } // VST1.16 <list>, [<Rn>]{!}, <Rm> (multiple single elements). void vst1_16(DRegisterList regs, MemOperand op, CoreRegister rm) { vst1(k16, regs, op, rm); } // VST1.16 <list>, [<Rn>]{!} (single element form one lane). void vst1_16(DRegisterLane dd, MemOperand op) { vst1(k16, dd, op); } // VST1.32 <list>, [<Rn>]{!} (multiple single elements). void vst1_32(DRegisterList regs, MemOperand op) { vst1(k32, regs, op); } // VST1.32 <list>, [<Rn>]{!}, <Rm> (multiple single elements). void vst1_32(DRegisterList regs, MemOperand op, CoreRegister rm) { vst1(k32, regs, op, rm); } // VST1.32 <list>, [<Rn>]{!} (single element form one lane). void vst1_32(DRegisterLane dd, MemOperand op) { vst1(k32, dd, op); } // VSTM <Rn>{!}, <list>, consecutive 64-bit registers. void vstm(MemOperand rn, DRegisterList regs); // VSTR <Sd>, [Rn{, #+/-<imm>}], store single extension register to memory. void vstr(SRegister rn, MemOperand op); // Binds Label l to the current location in the code buffer. void bind(Label& l); // Align the cursor to specified number of bytes, `n` must be a power of 2. void align(uint8_t n); private: void mov(Condition c, CoreRegister rd, CoreRegister rm); void b(Condition c, Label& l); void vdup(DataSize size, QRegister qd, DRegisterLane dm); void vmov_f32(Condition c, SRegister sd, SRegister sm); void vld1(DataSize size, DRegisterList regs, MemOperand op); void vld1(DataSize size, DRegisterList regs, MemOperand op, CoreRegister rm); void vst1(DataSize size, DRegisterList regs, MemOperand op); void vst1(DataSize size, DRegisterList regs, MemOperand op, CoreRegister rm); void vst1(DataSize size, DRegisterLane dd, MemOperand op); }; class MacroAssembler : public Assembler { using Assembler::Assembler; public: void f32_hardswish(QRegister sixth, QRegister three, QRegister six, QRegister zero, const QRegister *accs, size_t num_accs, const QRegister *tmps, size_t num_tmps); void Mov(CoreRegister rd, uint32_t imm); }; class TrampolineGenerator : public MacroAssembler { using MacroAssembler::MacroAssembler; public: void generate(size_t args_on_stack); private: // Helper functions to check that registers match. We keep the expected value inside of x0 and return early once we // have a mismatch. x0 then becomes the error code, if it is 0, there are no errors. void CheckRegisterMatch(DRegister actual, Label& exit); void CheckRegisterMatch(CoreRegister actual, Label& exit); }; } // namespace aarch32 } // namespace xnnpack
21,299
32.702532
118
h
XNNPACK
XNNPACK-master/src/xnnpack/aarch64-assembler.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <cstddef> #include <cstdint> #include <xnnpack/assembler.h> #include <xnnpack/microparams.h> namespace xnnpack { namespace aarch64 { constexpr size_t kInstructionSizeInBytesLog2 = 2; // Special values used to check that callee-saved registers are properly saved. // Low 8 bits should be 0 to encode register code. constexpr uint64_t kXRegisterCorruptValue = UINT64_C(0xDEADBEEF12345600); constexpr uint64_t kVRegisterCorruptValue = UINT64_C(0x7FF000007F801000); constexpr uint8_t kRegisterCorruptMask = UINT8_C(0xFF); // Instruction used to align code, is a hlt. constexpr uint32_t kAlignInstruction = 0xD4400000; struct WRegister { uint8_t code; }; constexpr WRegister w0{0}; constexpr WRegister w1{1}; constexpr WRegister w2{2}; constexpr WRegister w3{3}; constexpr WRegister w4{4}; constexpr WRegister w5{5}; constexpr WRegister w6{6}; constexpr WRegister w7{7}; constexpr WRegister w8{8}; constexpr WRegister w9{9}; constexpr WRegister w10{10}; constexpr WRegister w11{11}; constexpr WRegister w12{12}; constexpr WRegister w13{13}; constexpr WRegister w14{14}; constexpr WRegister w15{15}; constexpr WRegister w16{16}; constexpr WRegister w17{17}; constexpr WRegister w18{18}; constexpr WRegister w19{19}; constexpr WRegister w20{20}; constexpr WRegister w21{21}; constexpr WRegister w22{22}; constexpr WRegister w23{23}; constexpr WRegister w24{24}; constexpr WRegister w25{25}; constexpr WRegister w26{26}; constexpr WRegister w27{27}; constexpr WRegister w28{28}; constexpr WRegister w29{29}; constexpr WRegister w30{30}; struct XRegister { uint8_t code; }; constexpr XRegister x0{0}; constexpr XRegister x1{1}; constexpr XRegister x2{2}; constexpr XRegister x3{3}; constexpr XRegister x4{4}; constexpr XRegister x5{5}; constexpr XRegister x6{6}; constexpr XRegister x7{7}; constexpr XRegister x8{8}; constexpr XRegister x9{9}; constexpr XRegister x10{10}; constexpr XRegister x11{11}; constexpr XRegister x12{12}; constexpr XRegister x13{13}; constexpr XRegister x14{14}; constexpr XRegister x15{15}; constexpr XRegister x16{16}; constexpr XRegister x17{17}; constexpr XRegister x18{18}; constexpr XRegister x19{19}; constexpr XRegister x20{20}; constexpr XRegister x21{21}; constexpr XRegister x22{22}; constexpr XRegister x23{23}; constexpr XRegister x24{24}; constexpr XRegister x25{25}; constexpr XRegister x26{26}; constexpr XRegister x27{27}; constexpr XRegister x28{28}; constexpr XRegister x29{29}; constexpr XRegister x30{30}; constexpr XRegister xzr{31}; constexpr XRegister sp{31}; struct VRegisterLane { uint8_t code; uint8_t size; uint8_t lane; bool is_h() const { return size == 1; }; bool is_s() const { return size == 2; }; }; struct ScalarVRegister{ uint8_t code; uint8_t size; VRegisterLane operator[](std::size_t pos) const { return VRegisterLane{code, size, static_cast<uint8_t>(pos)}; } }; struct VRegister { uint8_t code; uint8_t size; uint8_t q; VRegister v8b() const { return {code, 0, 0}; } VRegister v16b() const { return {code, 0, 1}; } VRegister v4h() const { return {code, 1, 0}; } VRegister v8h() const { return {code, 1, 1}; } VRegister v2s() const { return {code, 2, 0}; } VRegister v4s() const { return {code, 2, 1}; } VRegister v1d() const { return {code, 3, 0}; } VRegister v2d() const { return {code, 3, 1}; } ScalarVRegister h() const { return {code, 1}; } ScalarVRegister s() const { return {code, 2}; } ScalarVRegister d() const { return {code, 3}; } bool is_h() const { return size == 1; }; bool is_s() const { return size == 2; }; }; constexpr VRegister v0{0}; constexpr VRegister v1{1}; constexpr VRegister v2{2}; constexpr VRegister v3{3}; constexpr VRegister v4{4}; constexpr VRegister v5{5}; constexpr VRegister v6{6}; constexpr VRegister v7{7}; constexpr VRegister v8{8}; constexpr VRegister v9{9}; constexpr VRegister v10{10}; constexpr VRegister v11{11}; constexpr VRegister v12{12}; constexpr VRegister v13{13}; constexpr VRegister v14{14}; constexpr VRegister v15{15}; constexpr VRegister v16{16}; constexpr VRegister v17{17}; constexpr VRegister v18{18}; constexpr VRegister v19{19}; constexpr VRegister v20{20}; constexpr VRegister v21{21}; constexpr VRegister v22{22}; constexpr VRegister v23{23}; constexpr VRegister v24{24}; constexpr VRegister v25{25}; constexpr VRegister v26{26}; constexpr VRegister v27{27}; constexpr VRegister v28{28}; constexpr VRegister v29{29}; constexpr VRegister v30{30}; constexpr VRegister v31{31}; struct VRegisterList { // NOLINTNEXTLINE(google-explicit-constructor) VRegisterList(VRegister vt1) : vt1(vt1), length(1) {} VRegisterList(VRegister vt1, VRegister vt2) : vt1(vt1), vt2(vt2), length(2) {} VRegisterList(VRegister vt1, VRegister vt2, VRegister vt3) : vt1(vt1), vt2(vt2), vt3(vt3), length(3) {} VRegisterList(VRegister vt1, VRegister vt2, VRegister vt3, VRegister vt4) : vt1(vt1), vt2(vt2), vt3(vt3), vt4(vt4), length(4) {} VRegister vt1; VRegister vt2; VRegister vt3; VRegister vt4; uint8_t length; }; struct ScalarVRegisterList { explicit ScalarVRegisterList(ScalarVRegister vt1) : vt1(vt1), length(1) {} ScalarVRegister vt1; uint8_t length; }; struct HRegister { uint8_t code; }; constexpr HRegister h0{0}; constexpr HRegister h1{1}; constexpr HRegister h2{2}; constexpr HRegister h3{3}; constexpr HRegister h4{4}; constexpr HRegister h5{5}; constexpr HRegister h6{6}; constexpr HRegister h7{7}; constexpr HRegister h8{8}; constexpr HRegister h9{9}; constexpr HRegister h10{10}; constexpr HRegister h11{11}; constexpr HRegister h12{12}; constexpr HRegister h13{13}; constexpr HRegister h14{14}; constexpr HRegister h15{15}; constexpr HRegister h16{16}; constexpr HRegister h17{17}; constexpr HRegister h18{18}; constexpr HRegister h19{19}; constexpr HRegister h20{20}; constexpr HRegister h21{21}; constexpr HRegister h22{22}; constexpr HRegister h23{23}; constexpr HRegister h24{24}; constexpr HRegister h25{25}; constexpr HRegister h26{26}; constexpr HRegister h27{27}; constexpr HRegister h28{28}; constexpr HRegister h29{29}; constexpr HRegister h30{30}; constexpr HRegister h31{31}; struct SRegister { uint8_t code; }; constexpr SRegister s0{0}; constexpr SRegister s1{1}; constexpr SRegister s2{2}; constexpr SRegister s3{3}; constexpr SRegister s4{4}; constexpr SRegister s5{5}; constexpr SRegister s6{6}; constexpr SRegister s7{7}; constexpr SRegister s8{8}; constexpr SRegister s9{9}; constexpr SRegister s10{10}; constexpr SRegister s11{11}; constexpr SRegister s12{12}; constexpr SRegister s13{13}; constexpr SRegister s14{14}; constexpr SRegister s15{15}; constexpr SRegister s16{16}; constexpr SRegister s17{17}; constexpr SRegister s18{18}; constexpr SRegister s19{19}; constexpr SRegister s20{20}; constexpr SRegister s21{21}; constexpr SRegister s22{22}; constexpr SRegister s23{23}; constexpr SRegister s24{24}; constexpr SRegister s25{25}; constexpr SRegister s26{26}; constexpr SRegister s27{27}; constexpr SRegister s28{28}; constexpr SRegister s29{29}; constexpr SRegister s30{30}; constexpr SRegister s31{31}; struct DRegister { uint8_t code; }; constexpr DRegister d0{0}; constexpr DRegister d1{1}; constexpr DRegister d2{2}; constexpr DRegister d3{3}; constexpr DRegister d4{4}; constexpr DRegister d5{5}; constexpr DRegister d6{6}; constexpr DRegister d7{7}; constexpr DRegister d8{8}; constexpr DRegister d9{9}; constexpr DRegister d10{10}; constexpr DRegister d11{11}; constexpr DRegister d12{12}; constexpr DRegister d13{13}; constexpr DRegister d14{14}; constexpr DRegister d15{15}; constexpr DRegister d16{16}; constexpr DRegister d17{17}; constexpr DRegister d18{18}; constexpr DRegister d19{19}; constexpr DRegister d20{20}; constexpr DRegister d21{21}; constexpr DRegister d22{22}; constexpr DRegister d23{23}; constexpr DRegister d24{24}; constexpr DRegister d25{25}; constexpr DRegister d26{26}; constexpr DRegister d27{27}; constexpr DRegister d28{28}; constexpr DRegister d29{29}; constexpr DRegister d30{30}; constexpr DRegister d31{31}; struct QRegister { uint8_t code; }; constexpr QRegister q0{0}; constexpr QRegister q1{1}; constexpr QRegister q2{2}; constexpr QRegister q3{3}; constexpr QRegister q4{4}; constexpr QRegister q5{5}; constexpr QRegister q6{6}; constexpr QRegister q7{7}; constexpr QRegister q8{8}; constexpr QRegister q9{9}; constexpr QRegister q10{10}; constexpr QRegister q11{11}; constexpr QRegister q12{12}; constexpr QRegister q13{13}; constexpr QRegister q14{14}; constexpr QRegister q15{15}; constexpr QRegister q16{16}; constexpr QRegister q17{17}; constexpr QRegister q18{18}; constexpr QRegister q19{19}; constexpr QRegister q20{20}; constexpr QRegister q21{21}; constexpr QRegister q22{22}; constexpr QRegister q23{23}; constexpr QRegister q24{24}; constexpr QRegister q25{25}; constexpr QRegister q26{26}; constexpr QRegister q27{27}; constexpr QRegister q28{28}; constexpr QRegister q29{29}; constexpr QRegister q30{30}; constexpr QRegister q31{31}; // C1.3.3 Load/Store addressing modes enum class AddressingMode { kOffset, // Base plus offset: [base{, #imm}] ; [base, Xm{, LSL #imm}]. kPostIndex, // Post-index: [base], #imm ; [base], Xm. kPreIndex, // Pre-index: [base, #imm]! }; struct MemOperand { // NOLINTNEXTLINE(google-explicit-constructor) MemOperand(XRegister xn): base(xn), mode(AddressingMode::kOffset), offset(0) {} MemOperand(XRegister xn, int32_t offset): base(xn), mode(AddressingMode::kOffset), offset(offset) {} MemOperand(XRegister xn, int32_t offset, AddressingMode mode): base(xn), mode(mode), offset(offset) {} // Overload postfix increment to indicate a pre-index addressing mode for load/stores. MemOperand operator++(int) { mode = AddressingMode::kPreIndex; return *this; } XRegister base; AddressingMode mode; int32_t offset; }; static inline MemOperand operator,(XRegister r, int32_t offset) { return MemOperand(r, offset); } // Helper struct for some syntax sugar to look like native assembly, see mem. struct MemOperandHelper { MemOperand operator[](MemOperand op) const { return op; } MemOperand operator[](XRegister r) const { return MemOperand(r, 0); } }; // Use "mem" (and its overload of array subscript operator) to get some syntax // that looks closer to native assembly when accessing memory. For example: // - ldp(x0, x1, mem[rn, offset]); // offset // - ldp(x0, x1, mem[rn], offset); // post-indexed constexpr MemOperandHelper mem; enum PrefetchOp { kPLDL1KEEP = 0, kPSTL1KEEP = 0x10, }; enum Condition : uint32_t { kEQ = 0x0, kNE = 0x1, kCS = 0x2, kCC = 0x3, kMI = 0x4, kPL = 0x5, kVS = 0x6, kVC = 0x7, kHI = 0x8, kLS = 0x9, kGE = 0xa, kLT = 0xB, kGT = 0xC, kLE = 0xD, kAL = 0xE, kHS = kCS, kLO = kCC, }; enum class BranchType { kConditional, // For encoding, TBZ and TBNZ are treated similarly, called TBXZ here. kTbxz, kUnconditional, }; // Instruction to use for alignment. // kNop should be used for loops, branch targets. kHlt for end of function. enum class AlignInstruction { kHlt, kNop, }; class Assembler : public AssemblerBase { public: using AssemblerBase::AssemblerBase; // Base instructions. void add(XRegister xd, XRegister xn, uint16_t imm12); void add(XRegister xd, XRegister xn, XRegister xm); void adds(XRegister xd, XRegister xn, uint16_t imm12); void ands(XRegister xd, XRegister xn, uint16_t imm12); void b(Label& l); void b_eq(Label& l) { return b(kEQ, l); } void b_hi(Label& l) { return b(kHI, l); } void b_hs(Label& l) { return b(kHS, l); } void b_lo(Label& l) { return b(kLO, l); } void b_ne(Label& l) { return b(kNE, l); } void bl(int32_t offset); void blr(XRegister xn); void cmp(XRegister xn, uint16_t imm12); void cmp(XRegister xn, XRegister xm); void csel(XRegister xd, XRegister xn, XRegister xm, Condition c); void hlt(); void ldp(XRegister xt1, XRegister xt2, MemOperand xn); void ldp(XRegister xt1, XRegister xt2, MemOperand xn, int32_t imm); void ldr(XRegister xt, MemOperand xn); void ldr(WRegister xt, MemOperand xn, int32_t imm); void ldr(XRegister xt, MemOperand xn, int32_t imm); void mov(XRegister xd, uint16_t imm); void mov(XRegister xd, XRegister xn); void movk(XRegister xd, uint16_t imm, uint8_t shift); void nop(); void prfm(PrefetchOp prfop, MemOperand xn); void ret(); void stp(XRegister xt1, XRegister xt2, MemOperand xn); void str(XRegister xt1, MemOperand xn); void sub(XRegister xd, XRegister xn, XRegister xm); void sub(XRegister xd, XRegister xn, uint16_t imm12); void subs(XRegister xd, XRegister xn, uint16_t imm12); void tbnz(XRegister xd, uint8_t bit, Label& l); void tbz(XRegister xd, uint8_t bit, Label& l); // Only immediates with lowest N bits set are supported. void tst(XRegister xn, uint8_t imm); // SIMD instructions void dup(DRegister vd, VRegisterLane vn); void dup(SRegister vd, VRegisterLane vn); void dup(VRegister vd, VRegisterLane vn); void fabs(VRegister vd, VRegister vn); void fadd(VRegister vd, VRegister vn, VRegister vm); void fmax(VRegister vd, VRegister vn, VRegister vm); void fmin(VRegister vd, VRegister vn, VRegister vm); void fmla(VRegister vd, VRegister vn, VRegisterLane vm); void fmul(VRegister vd, VRegister vn, VRegister vm); void fneg(VRegister vd, VRegister vn); void ins(VRegisterLane vd, XRegister vn); void ld1(VRegisterList vs, MemOperand xn, int32_t imm); // LD1 (single structure). // ld1({v1.d()}[0], mem[x0], 0) is invalid syntax, so the lane is a separate arg. void ld1(ScalarVRegisterList vs, size_t lane, MemOperand xn, int32_t imm); // Convenience overload for callers with single V register in list. void ld1(ScalarVRegister v, size_t lane, MemOperand xn, int32_t imm); void ld1r(VRegisterList xs, MemOperand xn); void ld2r(VRegisterList xs, MemOperand xn); void ld3r(VRegisterList xs, MemOperand xn); void ldp(DRegister dt1, DRegister dt2, MemOperand xn); void ldp(DRegister dt1, DRegister dt2, MemOperand xn, int32_t imm); void ldp(QRegister qt1, QRegister qt2, MemOperand xn, int32_t imm); void ldr(DRegister dt, MemOperand xn); void ldr(SRegister dt, MemOperand xn); void ldr(QRegister dt, MemOperand xn); void ldr(DRegister dt, MemOperand xn, int32_t imm); void ldr(HRegister dt, MemOperand xn, int32_t imm); void ldr(QRegister qt, MemOperand xn, int32_t imm); void ldr(SRegister st, MemOperand xn, int32_t imm); void mov(VRegister vd, VRegister vn); void movi(VRegister vd, uint8_t imm); // MOV (to general). void mov(XRegister xd, VRegisterLane vn); void st1(VRegisterList vs, MemOperand xn, int32_t imm); void st1(VRegisterList vs, MemOperand xn, XRegister xm); void stp(DRegister dt1, DRegister dt2, MemOperand xn); void stp(QRegister qt1, QRegister qt2, MemOperand xn); void stp(QRegister qt1, QRegister qt2, MemOperand xn, int32_t imm); void str(HRegister ht, MemOperand xn); void str(SRegister st, MemOperand xn); void str(DRegister dt, MemOperand xn, int32_t imm); void str(QRegister qt, MemOperand xn, int32_t imm); void str(SRegister st, MemOperand xn, int32_t imm); // Aligns the buffer to n (must be a power of 2). void align(uint8_t n, AlignInstruction instr); void align(uint8_t n) { align(n, AlignInstruction::kNop); } // Binds Label l to the current location in the code buffer. void bind(Label& l); private: void b(Condition c, Label& l); void branch_to_label(uint32_t opcode, BranchType bt, Label& l); void ld1_st1_multiple_structures(VRegisterList vs, MemOperand xn, int32_t imm, bool load); void ldr(uint32_t size, uint32_t opc, MemOperand xn, int32_t imm, uint8_t rt_code); void str(uint32_t size, uint32_t opc, MemOperand xn, int32_t imm, uint8_t rt_code); void tb_helper(uint32_t op, XRegister xd, uint8_t bit, Label& l); }; class MacroAssembler : public Assembler { using Assembler::Assembler; public: void f32_hardswish(VRegister sixth, VRegister three, VRegister six, VRegister zero, const VRegister *accs, size_t num_accs, const VRegister *tmps, size_t num_tmps); void Mov(XRegister xd, uint64_t imm); }; class TrampolineGenerator : public MacroAssembler { using MacroAssembler::MacroAssembler; public: void generate(size_t args_on_stack); private: // Helper functions to check that registers match. We keep the expected value inside of x0 and return early once we // have a mismatch. x0 then becomes the error code, if it is 0, there are no errors. void CheckRegisterMatch(VRegisterLane actual, Label& exit); void CheckRegisterMatch(XRegister actual, Label& exit); }; } // namespace aarch64 } // namespace xnnpack
16,880
29.471119
117
h
XNNPACK
XNNPACK-master/src/xnnpack/aligned-allocator.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <cstddef> #include <limits> #include <memory> #include <type_traits> #include <utility> #include <stdlib.h> #if defined(__ANDROID__) || defined(_WIN32) || defined(__CYGWIN__) #include <malloc.h> #endif template <typename T, size_t Alignment> class AlignedAllocator; template <size_t Alignment> class AlignedAllocator<void, Alignment> { public: typedef void* pointer; typedef const void* const_pointer; typedef void value_type; template <class U> struct rebind { typedef AlignedAllocator<U, Alignment> other; }; }; template <typename T, size_t Alignment> class AlignedAllocator { public: typedef T value_type; typedef T* pointer; typedef const T* const_pointer; typedef T& reference; typedef const T& const_reference; typedef size_t size_type; typedef ptrdiff_t difference_type; #if __cplusplus >= 201402L typedef std::true_type propagate_on_container_move_assignment; #endif template <class U> struct rebind { typedef AlignedAllocator<U, Alignment> other; }; public: inline AlignedAllocator() noexcept {} template <class U> inline AlignedAllocator( const AlignedAllocator<U, Alignment>& other) noexcept {} inline size_type max_size() const noexcept { return (std::numeric_limits<size_type>::max() - size_type(Alignment)) / sizeof(T); } inline pointer address(reference x) const noexcept { return std::addressof(x); } inline const_pointer address(const_reference x) const noexcept { return std::addressof(x); } inline pointer allocate( size_type n, typename AlignedAllocator<void, Alignment>::const_pointer hint = 0) { #if defined(_WIN32) void* memory = nullptr; memory = _aligned_malloc(n * sizeof(T), Alignment); if (memory == 0) { #if !defined(__GNUC__) && !defined(_MSC_VER) || defined(__EXCEPTIONS) || defined(_CPPUNWIND) throw std::bad_alloc(); #endif } #elif defined(__ANDROID__) || defined(__CYGWIN__) void* memory = memalign(Alignment, n * sizeof(T)); if (memory == 0) { #if !defined(__GNUC__) || defined(__EXCEPTIONS) throw std::bad_alloc(); #endif } #else void* memory = nullptr; if (posix_memalign(&memory, Alignment, n * sizeof(T)) != 0) { #if !defined(__GNUC__) || defined(__EXCEPTIONS) throw std::bad_alloc(); #endif } #endif return static_cast<pointer>(memory); } inline void deallocate(pointer p, size_type n) noexcept { #if defined(_WIN32) _aligned_free(static_cast<void*>(p)); #else free(static_cast<void*>(p)); #endif } template <class U, class... Args> inline void construct(U* p, Args&&... args) { ::new (static_cast<void*>(p)) U(std::forward<Args>(args)...); } template <class U> inline void destroy(U* p) { p->~U(); } };
2,998
22.992
92
h
XNNPACK
XNNPACK-master/src/xnnpack/allocator.h
// Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <limits.h> #include <stddef.h> #include <stdlib.h> #include <string.h> #if defined(_MSC_VER) #include <malloc.h> #elif !defined(__GNUC__) #include <alloca.h> #endif #include <xnnpack.h> #include <xnnpack/common.h> #include <xnnpack/params.h> #ifdef __cplusplus extern "C" { #endif XNN_INTERNAL extern const struct xnn_allocator xnn_default_allocator; inline static void* xnn_allocate_memory(size_t memory_size) { return xnn_params.allocator.allocate(xnn_params.allocator.context, memory_size); } inline static void* xnn_allocate_zero_memory(size_t memory_size) { void* memory_pointer = xnn_params.allocator.allocate(xnn_params.allocator.context, memory_size); if (memory_pointer != NULL) { memset(memory_pointer, 0, memory_size); } return memory_pointer; } inline static void* xnn_reallocate_memory(void* memory_pointer, size_t memory_size) { return xnn_params.allocator.reallocate(xnn_params.allocator.context, memory_pointer, memory_size); } inline static void xnn_release_memory(void* memory_pointer) { xnn_params.allocator.deallocate(xnn_params.allocator.context, memory_pointer); } inline static void* xnn_allocate_simd_memory(size_t memory_size) { return xnn_params.allocator.aligned_allocate(xnn_params.allocator.context, XNN_ALLOCATION_ALIGNMENT, memory_size); } inline static void* xnn_allocate_zero_simd_memory(size_t memory_size) { void* memory_pointer = xnn_params.allocator.aligned_allocate( xnn_params.allocator.context, XNN_ALLOCATION_ALIGNMENT, memory_size); if (memory_pointer != NULL) { memset(memory_pointer, 0, memory_size); } return memory_pointer; } inline static void xnn_release_simd_memory(void* memory_pointer) { xnn_params.allocator.aligned_deallocate(xnn_params.allocator.context, memory_pointer); } #if defined(__GNUC__) && defined(__BIGGEST_ALIGNMENT__) && (__BIGGEST_ALIGNMENT__ >= XNN_ALLOCATION_ALIGNMENT) #define XNN_SIMD_ALLOCA(size) __builtin_alloca((size)) #elif (defined(__clang_major__) && (__clang_major__ >= 4)) || \ (defined(__GNUC__) && (__GNUC__ >= 5 || __GNUC__ == 4 && __GNUC_MINOR__ >= 7) && !defined(__INTEL_COMPILER)) // Builtin expects alignment in bits. #define XNN_SIMD_ALLOCA(size) __builtin_alloca_with_align((size), XNN_ALLOCATION_ALIGNMENT * CHAR_BIT) #elif defined(__GNUC__) #define XNN_SIMD_ALLOCA(size) \ ((void*) ((((uintptr_t) __builtin_alloca((size) + XNN_ALLOCATION_ALIGNMENT)) | (XNN_ALLOCATION_ALIGNMENT - 1)) + 1)) #elif defined(_MSC_VER) #define XNN_SIMD_ALLOCA(size) \ ((void*) ((((uintptr_t) _alloca((size) + XNN_ALLOCATION_ALIGNMENT)) | (XNN_ALLOCATION_ALIGNMENT - 1)) + 1)) #else #define XNN_SIMD_ALLOCA(size) \ ((void*) ((((uintptr_t) alloca((size) + XNN_ALLOCATION_ALIGNMENT)) | (XNN_ALLOCATION_ALIGNMENT - 1)) + 1)) #endif #ifdef __cplusplus } // extern "C" #endif
3,012
33.632184
120
h
XNNPACK
XNNPACK-master/src/xnnpack/argmaxpool.h
// Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_F32_ARGMAXPOOL_UNIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_elements, \ size_t channels, \ const float** input, \ size_t input_offset, \ float* output, \ uint32_t* index, \ size_t input_increment, \ size_t output_increment); DECLARE_F32_ARGMAXPOOL_UNIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_4x__neon_c4) DECLARE_F32_ARGMAXPOOL_UNIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_4x__scalar_c1) DECLARE_F32_ARGMAXPOOL_UNIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_4x__sse2_c4) DECLARE_F32_ARGMAXPOOL_UNIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_4x__wasmsimd_c4) DECLARE_F32_ARGMAXPOOL_UNIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_9x__neon_c4) DECLARE_F32_ARGMAXPOOL_UNIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_9x__scalar_c1) DECLARE_F32_ARGMAXPOOL_UNIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_9x__sse2_c4) DECLARE_F32_ARGMAXPOOL_UNIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_9x__wasmsimd_c4) #define DECLARE_F32_ARGMAXPOOL_MULTIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_elements, \ size_t channels, \ const float** input, \ size_t input_offset, \ float* accumulation_buffer, \ uint32_t* index_buffer, \ float* output, \ uint32_t* index, \ size_t input_increment, \ size_t output_increment); DECLARE_F32_ARGMAXPOOL_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_9p8x__neon_c4) DECLARE_F32_ARGMAXPOOL_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_9p8x__scalar_c1) DECLARE_F32_ARGMAXPOOL_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_9p8x__sse2_c4) DECLARE_F32_ARGMAXPOOL_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_argmaxpool_ukernel_9p8x__wasmsimd_c4) #ifdef __cplusplus } // extern "C" #endif
2,992
45.765625
95
h
XNNPACK
XNNPACK-master/src/xnnpack/array-helpers.h
#pragma once #include <algorithm> #include <array> #include <cassert> #include <cstddef> namespace xnnpack { namespace internal { template <typename T, size_t N, typename F, size_t... Indx> void ArrayApplyImpl(std::array<T, N>&& args, F&& f, std::integer_sequence<size_t, Indx...> seq) { f(std::move(args[Indx])...); } template <typename T, size_t N, typename F, typename Indx = std::make_index_sequence<N>> void ArrayApply(std::array<T, N>&& args, F&& f) { return ArrayApplyImpl(std::move(args), f, Indx{}); } template <size_t... Is, typename V> constexpr std::array<V, sizeof...(Is)> MakeArrayImpl( V value, std::integer_sequence<size_t, Is...>) { return {((void)Is, value)...}; } template <size_t N, typename V> constexpr std::array<V, N> MakeArray(V value) { return MakeArrayImpl(value, std::make_index_sequence<N>{}); } template <typename T> static constexpr T kDefault{}; template <typename T, size_t max_size> class ArrayPrefix { public: constexpr ArrayPrefix(size_t size, T t) : size_(size), array_(MakeArray<max_size>(t)) { assert(size_ <= max_size); } explicit constexpr ArrayPrefix(size_t size) : size_(size) { assert(size_ <= max_size); } template <typename Array, typename = std::enable_if_t<!std::is_integral_v<Array>>> explicit constexpr ArrayPrefix(Array&& array) : ArrayPrefix({}) { for (const auto& v : array) { push_back(v); } } constexpr ArrayPrefix(std::initializer_list<T> init) : ArrayPrefix(init.size(), kDefault<T>) { assert(size_ <= max_size); std::copy(init.begin(), init.end(), begin()); } auto begin() { return array_.begin(); } auto begin() const { return array_.cbegin(); } auto end() { auto result = array_.begin(); std::advance(result, size_); return result; } auto end() const { auto result = array_.cbegin(); std::advance(result, size_); return result; } auto& operator[](size_t index) { assert(index < size_); return array_[index]; } const auto& operator[](size_t index) const { assert(index < size_); return array_[index]; } void push_back(const T& t) { assert(size_ + 1 <= max_size); array_[size_++] = t; } size_t size() const { return size_; } private: size_t size_; std::array<T, max_size> array_; }; } // namespace internal } // namespace xnnpack
2,396
24.5
68
h
XNNPACK
XNNPACK-master/src/xnnpack/assembler.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <xnnpack/memory.h> #include <array> #include <cstdint> typedef uint8_t byte; namespace xnnpack { constexpr size_t kInstructionSizeInBytes = 4; constexpr size_t kInstructionSizeInBytesLog2 = 2; enum class Error { kNoError, kOutOfMemory, kInvalidOperand, kLabelAlreadyBound, kLabelOffsetOutOfBounds, kLabelHasTooManyUsers, kInvalidLaneIndex, kInvalidRegisterListLength, kFinalizeCodeMemoryFail, kUnimplemented, }; // Biggest user of label is for callee-saved registers check in test mode. constexpr size_t max_label_users = 16; // Label is a target of a branch. You call Assembler::bind to bind a label to an // actual location in the instruction stream. // // ``` // Label l; // b(kAl, l1); // branch to an unbound label is fine, it will be patched later. // a.bind(l); // binds label to this location in the instruction stream. // b(kAl, l1); // branch to an already bound label. // ``` struct Label { // Location of label within Assembler buffer. byte* offset = nullptr; // A label can only be bound once, binding it again leads to an error. bool bound = (offset != nullptr); // All users of this label, recorded by their offset in the Assembler buffer. std::array<byte*, max_label_users> users{{0}}; size_t num_users = 0; // Records a user (e.g. branch instruction) of this label. // Returns true if success, false if number of users exceeds maximum. bool add_use(byte* offset) { if (num_users >= max_label_users) { return false; } users[num_users++] = offset; return true; } }; class AssemblerBase { public: // Takes an xnn_code_buffer with a pointer to allocated memory. If the buffer // already contains content (size != 0), appends to after size (up to capacity). explicit AssemblerBase(xnn_code_buffer* buf); // Write value into the code buffer and advances cursor_. void emit32(uint32_t value); void emit8(byte value); // Finish assembly of code, this should be the last function called on an // instance of Assembler. Returns a pointer to the start of code region. void* finalize(); // Reset the assembler state (no memory is freed). void reset(); // Get a pointer to the start of code buffer. const byte* start() const { return buffer_; } const byte* offset() const { return cursor_; } template<typename T> T offset() const { return reinterpret_cast<T>(cursor_); } // Returns the number of bytes of code actually in the buffer. size_t code_size_in_bytes() const { return (cursor_ - buffer_); } Error error() const { return error_; } protected: // Pointer into code buffer to start writing code. byte* buffer_; // Pointer to current position in code buffer. byte* cursor_; // Pointer to out-of-bounds of code buffer. byte* top_; // Errors encountered while assembling code. Error error_ = Error::kNoError; // Holds an xnn_code_buffer, will write code to its code pointer, and unmap // unused pages on finalizing. xnn_code_buffer* xnn_buffer = nullptr; private: template <typename Value> void emit(Value value) { if (error_ != Error::kNoError) { return; } if (cursor_ + sizeof(value) > top_) { error_ = Error::kOutOfMemory; return; } memcpy(cursor_, &value, sizeof(value)); cursor_ += sizeof(value); } }; } // namespace xnnpack
3,515
28.546218
82
h
XNNPACK
XNNPACK-master/src/xnnpack/avgpool.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #include <xnnpack/microparams.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_F16_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_elements, \ size_t channels, \ const void** input, \ size_t input_offset, \ const void* zero, \ void* buffer, \ void* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_f16_scaleminmax_params* params); DECLARE_F16_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_avgpool_minmax_ukernel_9p8x__f16c_c8) DECLARE_F16_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_avgpool_minmax_ukernel_9p8x__neonfp16arith_c8) #define DECLARE_F16_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_elements, \ size_t channels, \ const void** input, \ size_t input_offset, \ const void* zero, \ void* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_f16_scaleminmax_params* params); DECLARE_F16_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_avgpool_minmax_ukernel_9x__f16c_c8) DECLARE_F16_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_avgpool_minmax_ukernel_9x__neonfp16arith_c8) #define DECLARE_F32_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_elements, \ size_t channels, \ const float** input, \ size_t input_offset, \ const float* zero, \ float* buffer, \ float* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_f32_scaleminmax_params* params); DECLARE_F32_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9p8x__neon_c4) DECLARE_F32_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9p8x__scalar_c1) DECLARE_F32_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9p8x__sse_c4) DECLARE_F32_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9p8x__wasm_c1) DECLARE_F32_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9p8x__wasmsimd_arm_c4) DECLARE_F32_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9p8x__wasmsimd_x86_c4) #define DECLARE_F32_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_elements, \ size_t channels, \ const float** input, \ size_t input_offset, \ const float* zero, \ float* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_f32_scaleminmax_params* params); DECLARE_F32_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9x__neon_c4) DECLARE_F32_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9x__scalar_c1) DECLARE_F32_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9x__sse_c4) DECLARE_F32_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9x__wasm_c1) DECLARE_F32_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_arm_c4) DECLARE_F32_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_x86_c4) #define DECLARE_QU8_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_elements, \ size_t channels, \ const uint8_t** input, \ size_t input_offset, \ const uint8_t* zero, \ int32_t* buffer, \ uint8_t* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_qu8_avgpool_minmax_params* params); DECLARE_QU8_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_avgpool_minmax_fp32_ukernel_9p8x__neon_c8) DECLARE_QU8_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_avgpool_minmax_fp32_ukernel_9p8x__scalar_imagic_c1) DECLARE_QU8_AVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_avgpool_minmax_fp32_ukernel_9p8x__sse2_c8) #define DECLARE_QU8_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_elements, \ size_t channels, \ const uint8_t** input, \ size_t input_offset, \ const uint8_t* zero, \ uint8_t* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_qu8_avgpool_minmax_params* params); DECLARE_QU8_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_avgpool_minmax_fp32_ukernel_9x__neon_c8) DECLARE_QU8_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_avgpool_minmax_fp32_ukernel_9x__scalar_imagic_c1) DECLARE_QU8_AVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_avgpool_minmax_fp32_ukernel_9x__sse2_c8) #ifdef __cplusplus } // extern "C" #endif
8,026
56.335714
113
h
XNNPACK
XNNPACK-master/src/xnnpack/cache.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> // For size_t. #include <stdint.h> // For uint32_t. #include <xnnpack.h> // For xnn_status. #include <xnnpack/common.h> // For XNN_INLINE. #include <xnnpack/memory.h> // For xnn_code_buffer. #include <xnnpack/mutex.h> // For xnn_mutex. #ifdef __cplusplus extern "C" { #endif #define XNN_CACHE_NOT_FOUND SIZE_MAX // Return value when code is not found in the cache. // A cache for arbitrary bytes. // The implementation is similar to a hash table with open addressing and linear // probing, but restricted to our use cases. // Similar to buckets in a hash table implementation, this is an entry in the // cache. It stores "metadata" about the generated code (size and offset). The // actual bytes are in the cache's buffer. struct xnn_cache_bucket { // A hash for quick comparison. uint32_t hash; // Size of bytes. size_t size; // Offset of bytes, relative to cache's buffer. size_t offset; }; enum xnn_cache_type { xnn_cache_type_invalid = 0, xnn_cache_type_code, xnn_cache_type_weights, }; struct xnn_cache { enum xnn_cache_type type; // A growing buffer that is used to keep all generated code or repacked weights. union { struct xnn_code_buffer code; struct xnn_weights_buffer weights; }; // Entries in the cache. struct xnn_cache_bucket* buckets; // Capacity of the cache, when the load factor (num_entries/num_buckets) grows // beyond a limit, the cache is expanded. size_t num_buckets; size_t num_entries; size_t hits; size_t misses; }; // A cache for JIT generated microkernel code. struct xnn_code_cache { struct xnn_cache cache; }; enum xnn_status xnn_init_code_cache(struct xnn_code_cache* cache); enum xnn_status xnn_release_code_cache(struct xnn_code_cache* cache); // Looks up `ptr` in the cache, returns offset into cache's buffer if found. // `ptr` should already point into cache->buffer. // If it already exists within the cache, the buffer will be rewound, so we can // reuse the same section of the buffer. size_t xnn_get_or_insert_code_cache(struct xnn_code_cache* cache, void* ptr, size_t size); XNN_INLINE static bool xnn_code_cache_valid(struct xnn_code_cache* code_cache) { return code_cache != NULL && code_cache->cache.type == xnn_cache_type_code; } // The state of weights cache finalization. enum xnn_cache_state { // Not finalized. xnn_cache_state_not_finalized, // The underlying memory is trimmed to be as compact as possible. xnn_cache_state_hard_finalized, // The underlying memory has some extra space at the end. xnn_cache_state_soft_finalized, }; // A cache for repacked weights. struct xnn_weights_cache { struct xnn_cache cache; // Protects updates of `cache`, it has the same lifetime as `cache`, and so should be initialized/destroyed together // with the `cache`. struct xnn_mutex mutex; // Maximum size of packed weights that have been inserted into the cache. size_t max_weights_size; enum xnn_cache_state finalization_state; }; enum xnn_status xnn_init_weights_cache(struct xnn_weights_cache* cache); enum xnn_status xnn_init_weights_cache_with_size(struct xnn_weights_cache* cache, size_t size); // Finalizes the weights cache, so that we cannot insert any more entries into the cache. enum xnn_status xnn_finalize_weights_cache( struct xnn_weights_cache* cache, enum xnn_weights_cache_finalization_kind finalization_kind); enum xnn_status xnn_release_weights_cache(struct xnn_weights_cache* cache); // Ensures that cache has enough space for `n` bytes, locks the mutex to protect future updates. Mutex must be unlocked // using xnn_get_or_insert_weights_cache. void* xnn_reserve_space_in_weights_cache(struct xnn_weights_cache* cache, size_t n); // Looks up packed weights at `ptr` in the cache. If it is found, reuse it. Otherwise, it is added to the cache. Mutex // must already be locked before calling this, it will be unlocked at the end of this function. size_t xnn_get_or_insert_weights_cache(struct xnn_weights_cache* cache, void* ptr, size_t size); bool xnn_weights_cache_is_finalized(struct xnn_weights_cache* cache); #ifdef __cplusplus } // extern "C" #endif
4,355
36.230769
119
h
XNNPACK
XNNPACK-master/src/xnnpack/common.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #if defined(__APPLE__) #include <TargetConditionals.h> #endif // Define architecture identification macros #if defined(__i386__) || defined(__i486__) || defined(__i586__) || defined(__i686__) || defined(_M_IX86) #define XNN_ARCH_X86 1 #else #define XNN_ARCH_X86 0 #endif #if defined(__x86_64__) || defined(__x86_64) || defined(_M_X64) && !defined(_M_ARM64EC) #define XNN_ARCH_X86_64 1 #else #define XNN_ARCH_X86_64 0 #endif #if defined(__arm__) || defined(_M_ARM) #define XNN_ARCH_ARM 1 #else #define XNN_ARCH_ARM 0 #endif #if defined(__aarch64__) || defined(_M_ARM64) || defined(_M_ARM64EC) #define XNN_ARCH_ARM64 1 #else #define XNN_ARCH_ARM64 0 #endif #if defined(__PPC64__) || defined(__ppc64__) || defined(__powerpc64__) || defined(_ARCH_PPC64) #define XNN_ARCH_PPC64 1 #else #define XNN_ARCH_PPC64 0 #endif #if defined(__riscv) || defined(__riscv__) #define XNN_ARCH_RISCV 1 #else #define XNN_ARCH_RISCV 0 #endif #if defined(__hexagon__) #define XNN_ARCH_HEXAGON 1 #else #define XNN_ARCH_HEXAGON 0 #endif #if defined(__wasm__) #if defined(__wasm_relaxed_simd__) #define XNN_ARCH_WASM 0 #define XNN_ARCH_WASMSIMD 0 #define XNN_ARCH_WASMRELAXEDSIMD 1 #elif defined(__wasm_simd128__) #define XNN_ARCH_WASM 0 #define XNN_ARCH_WASMSIMD 1 #define XNN_ARCH_WASMRELAXEDSIMD 0 #else #define XNN_ARCH_WASM 1 #define XNN_ARCH_WASMSIMD 0 #define XNN_ARCH_WASMRELAXEDSIMD 0 #endif #else #define XNN_ARCH_WASM 0 #define XNN_ARCH_WASMSIMD 0 #define XNN_ARCH_WASMRELAXEDSIMD 0 #endif // Define platform identification macros #if defined(__ANDROID__) #define XNN_PLATFORM_ANDROID 1 #else #define XNN_PLATFORM_ANDROID 0 #endif #if defined(__linux__) #define XNN_PLATFORM_LINUX 1 #else #define XNN_PLATFORM_LINUX 0 #endif #if defined(__APPLE__) && TARGET_OS_IPHONE // iOS on iPhone / iPad Touch, iPad OS, watchOS, or tvOS #define XNN_PLATFORM_IOS 1 #else #define XNN_PLATFORM_IOS 0 #endif #if defined(__APPLE__) && TARGET_OS_MAC #define XNN_PLATFORM_MAC 1 #else #define XNN_PLATFORM_MAC 0 #endif #if XNN_PLATFORM_ANDROID || XNN_PLATFORM_IOS #define XNN_PLATFORM_MOBILE 1 #else #define XNN_PLATFORM_MOBILE 0 #endif #if defined(__EMSCRIPTEN__) || defined(__wasm__) #define XNN_PLATFORM_WEB 1 #else #define XNN_PLATFORM_WEB 0 #endif #if defined(_WIN32) #define XNN_PLATFORM_WINDOWS 1 #else #define XNN_PLATFORM_WINDOWS 0 #endif #if defined(__Fuchsia__) #define XNN_PLATFORM_FUCHSIA 1 #else #define XNN_PLATFORM_FUCHSIA 0 #endif #if defined(__hexagon__) && !defined(__linux__) #define XNN_PLATFORM_QURT 1 #else #define XNN_PLATFORM_QURT 0 #endif #ifndef XNN_PLATFORM_JIT #if (XNN_ARCH_ARM || XNN_ARCH_ARM64) && !XNN_PLATFORM_IOS && !XNN_PLATFORM_FUCHSIA || XNN_PLATFORM_WEB #define XNN_PLATFORM_JIT 1 #else #define XNN_PLATFORM_JIT 0 #endif #endif // XNN_PLATFORM_JIT // Define compile identification macros #if defined(__clang__) #define XNN_COMPILER_CLANG 1 #elif defined(__INTEL_COMPILER) #define XNN_COMPILER_ICC 1 #elif defined(_MSC_VER) #define XNN_COMPILER_MSVC 1 #elif defined(__GNUC__) #define XNN_COMPILER_GCC 1 #endif #ifndef XNN_COMPILER_CLANG #define XNN_COMPILER_CLANG 0 #endif #ifndef XNN_COMPILER_GCC #define XNN_COMPILER_GCC 0 #endif #ifndef XNN_COMPILER_MSVC #define XNN_COMPILER_MSVC 0 #endif #ifndef XNN_COMPILER_ICC #define XNN_COMPILER_ICC 0 #endif #ifndef XNN_TEST_MODE #define XNN_TEST_MODE 0 #endif #ifndef XNN_MAX_UARCH_TYPES #if (XNN_ARCH_ARM || XNN_ARCH_ARM64) && !XNN_PLATFORM_IOS #define XNN_MAX_UARCH_TYPES 3 #else #define XNN_MAX_UARCH_TYPES 1 #endif #endif #define XNN_UARCH_DEFAULT 0 #if defined(__has_builtin) #define XNN_COMPILER_HAS_BUILTIN(builtin) __has_builtin(builtin) #else #define XNN_COMPILER_HAS_BUILTIN(builtin) 0 #endif #if defined(__has_feature) #define XNN_COMPILER_HAS_FEATURE(builtin) __has_feature(builtin) #else #define XNN_COMPILER_HAS_FEATURE(builtin) 0 #endif #if defined(__GNUC__) #if defined(__clang__) || (__GNUC__ > 4 || __GNUC__ == 4 && __GNUC_MINOR__ >= 5) #define XNN_UNREACHABLE do { __builtin_unreachable(); } while (0) #else #define XNN_UNREACHABLE do { __builtin_trap(); } while (0) #endif #elif defined(_MSC_VER) #define XNN_UNREACHABLE __assume(0) #else #define XNN_UNREACHABLE do { } while (0) #endif #if defined(__GNUC__) #define XNN_ALIGN(alignment) __attribute__((__aligned__(alignment))) #elif defined(_MSC_VER) #define XNN_ALIGN(alignment) __declspec(align(alignment)) #else #error "Platform-specific implementation of XNN_ALIGN required" #endif #if defined(__GNUC__) #define XNN_UNALIGNED __attribute__((__aligned__(1))) #elif defined(_MSC_VER) #if defined(_M_IX86) #define XNN_UNALIGNED #else #define XNN_UNALIGNED __unaligned #endif #else #error "Platform-specific implementation of XNN_UNALIGNED required" #endif #define XNN_COUNT_OF(array) (sizeof(array) / sizeof(0[array])) #if defined(__cplusplus) || XNN_COMPILER_MSVC || XNN_COMPILER_CLANG // static as array indices in function parameter declaration is a C99 feature, not supported in C++. // MSVC does not support this feature, even in C mode. // Clang generates suboptimal code, see https://github.com/llvm/llvm-project/issues/59120 #define XNN_MIN_ELEMENTS(count) count #else #define XNN_MIN_ELEMENTS(count) static count #endif #if defined(__GNUC__) #define XNN_LIKELY(condition) (__builtin_expect(!!(condition), 1)) #define XNN_UNLIKELY(condition) (__builtin_expect(!!(condition), 0)) #else #define XNN_LIKELY(condition) (!!(condition)) #define XNN_UNLIKELY(condition) (!!(condition)) #endif #if XNN_COMPILER_HAS_BUILTIN(__builtin_unpredictable) #define XNN_UNPREDICTABLE(condition) (__builtin_unpredictable(!!(condition))) #elif defined(__GNUC__) && (__GNUC__ >= 9) && !defined(__INTEL_COMPILER) #define XNN_UNPREDICTABLE(condition) (__builtin_expect_with_probability(!!(condition), 0, 0.5)) #else #define XNN_UNPREDICTABLE(condition) (!!(condition)) #endif #if XNN_COMPILER_HAS_FEATURE(thread_sanitizer) #define XNN_DISABLE_TSAN __attribute__((__no_sanitize__("thread"))) #else #define XNN_DISABLE_TSAN #endif #if XNN_COMPILER_HAS_FEATURE(memory_sanitizer) #define XNN_DISABLE_MSAN __attribute__((__no_sanitize__("memory"))) #else #define XNN_DISABLE_MSAN #endif #if XNN_COMPILER_HAS_FEATURE(hwaddress_sanitizer) #define XNN_DISABLE_HWASAN __attribute__((__no_sanitize__("hwaddress"))) #else #define XNN_DISABLE_HWASAN #endif #define XNN_OOB_READS XNN_DISABLE_TSAN XNN_DISABLE_MSAN XNN_DISABLE_HWASAN #if defined(__GNUC__) #define XNN_INTRINSIC inline __attribute__((__always_inline__, __artificial__)) #elif defined(_MSC_VER) #define XNN_INTRINSIC __forceinline #else #define XNN_INTRINSIC inline #endif #if defined(__GNUC__) #define XNN_INLINE inline __attribute__((__always_inline__)) #elif defined(_MSC_VER) #define XNN_INLINE __forceinline #else #define XNN_INLINE inline #endif #ifndef XNN_INTERNAL #if defined(__ELF__) #define XNN_INTERNAL __attribute__((__visibility__("internal"))) #elif defined(__MACH__) #define XNN_INTERNAL __attribute__((__visibility__("hidden"))) #else #define XNN_INTERNAL #endif #endif #ifndef XNN_PRIVATE #if defined(__ELF__) #define XNN_PRIVATE __attribute__((__visibility__("hidden"))) #elif defined(__MACH__) #define XNN_PRIVATE __attribute__((__visibility__("hidden"))) #else #define XNN_PRIVATE #endif #endif #if defined(__clang__) #define XNN_PRAGMA_CLANG(pragma) _Pragma(pragma) #else #define XNN_PRAGMA_CLANG(pragma) #endif #if XNN_ARCH_WASM #define XNN_ALLOCATION_ALIGNMENT 4 #elif XNN_ARCH_X86 || XNN_ARCH_X86_64 #if XNN_PLATFORM_MOBILE #define XNN_ALLOCATION_ALIGNMENT 32 #else #define XNN_ALLOCATION_ALIGNMENT 64 #endif #else #define XNN_ALLOCATION_ALIGNMENT 16 #endif // Number of extra elements to allocate for DWCONV accumulators/buffers. #if XNN_ARCH_X86 || XNN_ARCH_X86_64 // For AVX512. #define XNN_MAX_SIMD_SIZE 64 #elif XNN_ARCH_RISCV || XNN_ARCH_WASM // Scalable vectors, assume masked loads and stores. // Wasm without SIMD. #define XNN_MAX_SIMD_SIZE 0 #elif XNN_ARCH_HEXAGON #define XNN_MAX_SIMD_SIZE 128 #else // XNN_ARCH_ARM, XNN_ARCH_ARM64, XNN_ARCH_WASMSIMD, XNN_ARCH_WASMRELAXEDSIMD. #define XNN_MAX_SIMD_SIZE 16 #endif // Use constant here to avoid dependency on xnnpack.h #if XNN_MAX_SIMD_SIZE >= 16 #define XNN_MULTIPASS_EXTRA_BYTES XNN_MAX_SIMD_SIZE #else #define XNN_MULTIPASS_EXTRA_BYTES 16 #endif #define XNN_LOG2_SIZEOF_INT8_T 0 // log2(sizeof(int8_t)) #define XNN_LOG2_SIZEOF_UINT8_T 0 // log2(sizeof(uint8_t)) #define XNN_LOG2_SIZEOF_INT16_T 1 // log2(sizeof(int16_t)) #define XNN_LOG2_SIZEOF_UINT16_T 1 // log2(sizeof(uint16_t)) #define XNN_LOG2_SIZEOF_HALF 1 // log2(sizeof(half)) #define XNN_LOG2_SIZEOF_FLOAT 2 // log2(sizeof(float)) #define XNN_LOG2_SIZEOF_INT32_T 2 // log2(sizeof(int32_t)) #define XNN_LOG2_SIZEOF_UINT32_T 2 // log2(sizeof(uint32_t))
9,330
24.634615
104
h
XNNPACK
XNNPACK-master/src/xnnpack/conv.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #include <xnnpack/microparams.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t input_height, \ size_t input_width, \ size_t output_y_start, \ size_t output_y_end, \ const float* input, \ const float* zero, \ const float* weights, \ float* output, \ size_t input_padding_top, \ size_t output_channels, \ size_t output_height_stride, \ size_t output_width_stride, \ const union xnn_f32_minmax_params* params); DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p1c3x4__aarch64_neonfma_2x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p1c3x4__aarch64_neonfma_2x2) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p1c3x4__neon_2x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p1c3x4__neon_2x2) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p1c3x4__scalar_1x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p1c3x8__aarch64_neonfma_2x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p1c3x8__aarch64_neonfma_2x2) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p1c3x8__neon_2x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p1c3x8__neon_2x2) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p0p1c3x4__aarch64_neonfma_2x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p0p1c3x4__aarch64_neonfma_2x2) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p0p1c3x4__neon_2x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p0p1c3x4__neon_2x2) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p0p1c3x4__scalar_1x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p0p1c3x8__aarch64_neonfma_2x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p0p1c3x8__aarch64_neonfma_2x2) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p0p1c3x8__neon_2x1) DECLARE_F32_CONV_HWC_UKERNEL_FUNCTION(xnn_f32_conv_hwc_ukernel_3x3s2p0p1c3x8__neon_2x2) #define DECLARE_F32_CONV_HWC2CHW_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t input_height, \ size_t input_width, \ size_t output_y_start, \ size_t output_y_end, \ const float* input, \ const float* zero, \ const float* weights, \ float* output, \ size_t input_padding_top, \ size_t output_channels, \ size_t output_height_stride, \ size_t output_channel_stride, \ const union xnn_f32_minmax_params* params); DECLARE_F32_CONV_HWC2CHW_UKERNEL_FUNCTION(xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__aarch64_neonfma_2x2) DECLARE_F32_CONV_HWC2CHW_UKERNEL_FUNCTION(xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__neon_2x2) DECLARE_F32_CONV_HWC2CHW_UKERNEL_FUNCTION(xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__scalar_1x1) DECLARE_F32_CONV_HWC2CHW_UKERNEL_FUNCTION(xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__sse_1x1) DECLARE_F32_CONV_HWC2CHW_UKERNEL_FUNCTION(xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__sse_2x2) DECLARE_F32_CONV_HWC2CHW_UKERNEL_FUNCTION(xnn_f32_conv_hwc2chw_ukernel_3x3s2p1c3x4__wasmsimd_2x2) #define DECLARE_F16_CONV_HWC2CHW_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t input_height, \ size_t input_width, \ size_t output_y_start, \ size_t output_y_end, \ const void* input, \ const void* zero, \ const void* weights, \ void* output, \ size_t input_padding_top, \ size_t output_channels, \ size_t output_height_stride, \ size_t output_channel_stride, \ const union xnn_f16_minmax_params* params); DECLARE_F16_CONV_HWC2CHW_UKERNEL_FUNCTION(xnn_f16_conv_hwc2chw_ukernel_3x3s2p1c3x4__neonfp16arith_2x2) #ifdef __cplusplus } // extern "C" #endif
5,501
52.417476
104
h
XNNPACK
XNNPACK-master/src/xnnpack/fft.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t batch, \ size_t samples, \ int16_t* data, \ const int16_t* twiddle, \ size_t stride); DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_samples1_ukernel__asm_aarch32_neon_x1) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_samples1_ukernel__asm_aarch32_neon_x2) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_samples1_ukernel__asm_aarch32_neon_x4) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_samples1_ukernel__neon) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_samples1_ukernel__scalar) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_samples4_ukernel__neon) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_samples4_ukernel__scalar) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_ukernel__neon_x1) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_ukernel__neon_x4) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_ukernel__scalar_x1) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_ukernel__scalar_x2) DECLARE_CS16_BFLY4_UKERNEL_FUNCTION(xnn_cs16_bfly4_ukernel__scalar_x4) #define DECLARE_CS16_FFTR_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t samples, \ int16_t* data, \ const int16_t* twiddle); DECLARE_CS16_FFTR_UKERNEL_FUNCTION(xnn_cs16_fftr_ukernel__asm_aarch32_neon_x1) DECLARE_CS16_FFTR_UKERNEL_FUNCTION(xnn_cs16_fftr_ukernel__asm_aarch32_neon_x4) DECLARE_CS16_FFTR_UKERNEL_FUNCTION(xnn_cs16_fftr_ukernel__neon_x4) DECLARE_CS16_FFTR_UKERNEL_FUNCTION(xnn_cs16_fftr_ukernel__scalar_x1) DECLARE_CS16_FFTR_UKERNEL_FUNCTION(xnn_cs16_fftr_ukernel__scalar_x2) DECLARE_CS16_FFTR_UKERNEL_FUNCTION(xnn_cs16_fftr_ukernel__scalar_x4) #ifdef __cplusplus } // extern "C" #endif
2,289
40.636364
89
h
XNNPACK
XNNPACK-master/src/xnnpack/fill.h
// Copyright 2020 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_FILL_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t kernel_elements, \ size_t channels, \ void* output, \ size_t output_stride, \ const uint32_t fill_pattern); DECLARE_FILL_UKERNEL_FUNCTION(xnn_xx_fill_ukernel__neon_x64) DECLARE_FILL_UKERNEL_FUNCTION(xnn_xx_fill_ukernel__scalar_x16) DECLARE_FILL_UKERNEL_FUNCTION(xnn_xx_fill_ukernel__sse2_x64) DECLARE_FILL_UKERNEL_FUNCTION(xnn_xx_fill_ukernel__wasmsimd_x64) #ifdef __cplusplus } // extern "C" #endif
914
25.142857
72
h
XNNPACK
XNNPACK-master/src/xnnpack/filterbank.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_U32_FILTERBANK_ACCUMULATE_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t rows, \ const uint32_t* input, \ const uint8_t* weight_widths, \ const uint16_t* weights, \ uint64_t* output); DECLARE_U32_FILTERBANK_ACCUMULATE_UKERNEL_FUNCTION(xnn_u32_filterbank_accumulate_ukernel__asm_aarch32_arm_x1) DECLARE_U32_FILTERBANK_ACCUMULATE_UKERNEL_FUNCTION(xnn_u32_filterbank_accumulate_ukernel__asm_aarch32_neon_x1) DECLARE_U32_FILTERBANK_ACCUMULATE_UKERNEL_FUNCTION(xnn_u32_filterbank_accumulate_ukernel__asm_aarch32_neon_x2) DECLARE_U32_FILTERBANK_ACCUMULATE_UKERNEL_FUNCTION(xnn_u32_filterbank_accumulate_ukernel__neon_x1) DECLARE_U32_FILTERBANK_ACCUMULATE_UKERNEL_FUNCTION(xnn_u32_filterbank_accumulate_ukernel__neon_x2) DECLARE_U32_FILTERBANK_ACCUMULATE_UKERNEL_FUNCTION(xnn_u32_filterbank_accumulate_ukernel__scalar_x1) #define DECLARE_U32_FILTERBANK_SUBTRACT_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t batch_size, \ const uint32_t* input, \ uint32_t smoothing, \ uint32_t alternate_smoothing, \ uint32_t one_minus_smoothing, \ uint32_t alternate_one_minus_smoothing, \ uint32_t min_signal_remaining, \ uint32_t smoothing_bits, \ uint32_t spectral_subtraction_bits, \ uint32_t* noise_estimate, \ uint32_t* output); DECLARE_U32_FILTERBANK_SUBTRACT_UKERNEL_FUNCTION(xnn_u32_filterbank_subtract_ukernel__scalar_x2) #ifdef __cplusplus } // extern "C" #endif
2,349
42.518519
110
h
XNNPACK
XNNPACK-master/src/xnnpack/gavgpool.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #include <xnnpack/microparams.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_F32_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t rows, \ size_t channels, \ const float* input, \ size_t input_stride, \ const float* zero, \ float* buffer, \ float* output, \ const union xnn_f32_scaleminmax_params* params); DECLARE_F32_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7p7x__neon_c4) DECLARE_F32_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7p7x__scalar_c1) DECLARE_F32_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7p7x__sse_c4) DECLARE_F32_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7p7x__wasm_c1) DECLARE_F32_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7p7x__wasmsimd_arm_c4) DECLARE_F32_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7p7x__wasmsimd_x86_c4) #define DECLARE_F32_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t rows, \ size_t channels, \ const float* input, \ size_t input_stride, \ const float* zero, \ float* output, \ const union xnn_f32_scaleminmax_params* params); DECLARE_F32_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7x__neon_c4) DECLARE_F32_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7x__scalar_c1) DECLARE_F32_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7x__sse_c4) DECLARE_F32_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7x__wasm_c1) DECLARE_F32_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7x__wasmsimd_arm_c4) DECLARE_F32_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f32_gavgpool_minmax_ukernel_7x__wasmsimd_x86_c4) #define DECLARE_F16_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t rows, \ size_t channels, \ const void* input, \ size_t input_stride, \ const void* zero, \ void* buffer, \ void* output, \ const union xnn_f16_scaleminmax_params* params); DECLARE_F16_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7p7x__neonfp16arith_c8) DECLARE_F16_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7p7x__neonfp16arith_c16) DECLARE_F16_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7p7x__neonfp16arith_c24) DECLARE_F16_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7p7x__neonfp16arith_c32) DECLARE_F16_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7p7x__f16c_c8) DECLARE_F16_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7p7x__f16c_c16) DECLARE_F16_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7p7x__f16c_c24) DECLARE_F16_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7p7x__f16c_c32) #define DECLARE_F16_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t rows, \ size_t channels, \ const void* input, \ size_t input_stride, \ const void* zero, \ void* output, \ const union xnn_f16_scaleminmax_params* params); DECLARE_F16_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7x__neonfp16arith_c8) DECLARE_F16_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7x__neonfp16arith_c16) DECLARE_F16_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7x__neonfp16arith_c24) DECLARE_F16_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7x__neonfp16arith_c32) DECLARE_F16_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7x__f16c_c8) DECLARE_F16_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7x__f16c_c16) DECLARE_F16_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7x__f16c_c24) DECLARE_F16_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_f16_gavgpool_minmax_ukernel_7x__f16c_c32) #define DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t rows, \ size_t channels, \ const int8_t* input, \ size_t input_stride, \ const int8_t* zero, \ int32_t* buffer, \ int8_t* output, \ const union xnn_qs8_avgpool_minmax_params* params); DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__neon_c8) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__neon_c16) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__neon_c24) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__neon_c32) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_rndnu_ukernel_7p7x__neon_c8) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_rndnu_ukernel_7p7x__neon_c16) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_rndnu_ukernel_7p7x__neon_c24) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_rndnu_ukernel_7p7x__neon_c32) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__neonv8_c8) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__neonv8_c16) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__neonv8_c24) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__neonv8_c32) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__sse2_c8) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__sse2_c16) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__sse2_c24) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__sse41_c8) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__sse41_c16) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__sse41_c24) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__wasmsimd_c8) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__wasmsimd_c16) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__wasmsimd_c24) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__wasmsimd_c32) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_imagic_c1) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_imagic_c2) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_imagic_c4) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_fmagic_c1) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_fmagic_c2) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_fmagic_c4) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_lrintf_c1) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_lrintf_c2) DECLARE_QS8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_lrintf_c4) #define DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t rows, \ size_t channels, \ const int8_t* input, \ size_t input_stride, \ const int8_t* zero, \ int8_t* output, \ const union xnn_qs8_avgpool_minmax_params* params); DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__neon_c8) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__neon_c16) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__neon_c24) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__neon_c32) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_rndnu_ukernel_7x__neon_c8) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_rndnu_ukernel_7x__neon_c16) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_rndnu_ukernel_7x__neon_c24) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_rndnu_ukernel_7x__neon_c32) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__neonv8_c8) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__neonv8_c16) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__neonv8_c24) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__neonv8_c32) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__sse2_c8) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__sse2_c16) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__sse2_c24) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__sse41_c8) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__sse41_c16) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__sse41_c24) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__wasmsimd_c8) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__wasmsimd_c16) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__wasmsimd_c24) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__wasmsimd_c32) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__scalar_imagic_c1) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__scalar_imagic_c2) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__scalar_imagic_c4) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__scalar_fmagic_c1) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__scalar_fmagic_c2) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__scalar_fmagic_c4) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__scalar_lrintf_c1) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__scalar_lrintf_c2) DECLARE_QS8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qs8_gavgpool_minmax_fp32_ukernel_7x__scalar_lrintf_c4) #define DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t rows, \ size_t channels, \ const uint8_t* input, \ size_t input_stride, \ const uint8_t* zero, \ int32_t* buffer, \ uint8_t* output, \ const union xnn_qu8_avgpool_minmax_params* params); DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__neon_c8) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__neon_c16) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__neon_c24) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__neon_c32) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_rndnu_ukernel_7p7x__neon_c8) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_rndnu_ukernel_7p7x__neon_c16) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_rndnu_ukernel_7p7x__neon_c24) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_rndnu_ukernel_7p7x__neon_c32) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__neonv8_c8) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__neonv8_c16) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__neonv8_c24) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__neonv8_c32) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__sse2_c8) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__sse2_c16) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__sse2_c24) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__sse41_c8) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__sse41_c16) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__sse41_c24) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__wasmsimd_c8) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__wasmsimd_c16) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__wasmsimd_c24) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__wasmsimd_c32) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_imagic_c1) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_imagic_c2) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_imagic_c4) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_fmagic_c1) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_fmagic_c2) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_fmagic_c4) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_lrintf_c1) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_lrintf_c2) DECLARE_QU8_GAVGPOOL_MINMAX_MULTIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7p7x__scalar_lrintf_c4) #define DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t rows, \ size_t channels, \ const uint8_t* input, \ size_t input_stride, \ const uint8_t* zero, \ uint8_t* output, \ const union xnn_qu8_avgpool_minmax_params* params); DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__neon_c8) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__neon_c16) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__neon_c24) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__neon_c32) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_rndnu_ukernel_7x__neon_c8) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_rndnu_ukernel_7x__neon_c16) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_rndnu_ukernel_7x__neon_c24) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_rndnu_ukernel_7x__neon_c32) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__neonv8_c8) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__neonv8_c16) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__neonv8_c24) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__neonv8_c32) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__sse2_c8) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__sse2_c16) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__sse2_c24) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__sse41_c8) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__sse41_c16) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__sse41_c24) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__wasmsimd_c8) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__wasmsimd_c16) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__wasmsimd_c24) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__wasmsimd_c32) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__scalar_imagic_c1) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__scalar_imagic_c2) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__scalar_imagic_c4) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__scalar_fmagic_c1) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__scalar_fmagic_c2) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__scalar_fmagic_c4) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__scalar_lrintf_c1) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__scalar_lrintf_c2) DECLARE_QU8_GAVGPOOL_MINMAX_UNIPASS_UKERNEL_FUNCTION(xnn_qu8_gavgpool_minmax_fp32_ukernel_7x__scalar_lrintf_c4) #define DECLARE_F32_GAVGPOOL_CW_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t elements, \ size_t channels, \ const float* input, \ float* output, \ const union xnn_f32_gavgpool_params* params); DECLARE_F32_GAVGPOOL_CW_UKERNEL_FUNCTION(xnn_f32_gavgpool_cw_ukernel__neon_x4) DECLARE_F32_GAVGPOOL_CW_UKERNEL_FUNCTION(xnn_f32_gavgpool_cw_ukernel__scalar_x1) DECLARE_F32_GAVGPOOL_CW_UKERNEL_FUNCTION(xnn_f32_gavgpool_cw_ukernel__sse_x4) DECLARE_F32_GAVGPOOL_CW_UKERNEL_FUNCTION(xnn_f32_gavgpool_cw_ukernel__wasmsimd_arm_x4) DECLARE_F32_GAVGPOOL_CW_UKERNEL_FUNCTION(xnn_f32_gavgpool_cw_ukernel__wasmsimd_x86_x4) #define DECLARE_F16_GAVGPOOL_CW_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t elements, \ size_t channels, \ const void* input, \ void* output, \ const union xnn_f16_gavgpool_params* params); DECLARE_F16_GAVGPOOL_CW_UKERNEL_FUNCTION(xnn_f16_gavgpool_cw_ukernel__neonfp16arith_x8) #ifdef __cplusplus } // extern "C" #endif
23,703
69.338279
115
h
XNNPACK
XNNPACK-master/src/xnnpack/ibilinear.h
// Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_F16_IBILINEAR_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t channels, \ const void** input, \ size_t input_offset, \ const void* weights, \ void* output, \ size_t output_increment); DECLARE_F16_IBILINEAR_UKERNEL_FUNCTION(xnn_f16_ibilinear_ukernel__fma3_c8) DECLARE_F16_IBILINEAR_UKERNEL_FUNCTION(xnn_f16_ibilinear_ukernel__fma3_c16) DECLARE_F16_IBILINEAR_UKERNEL_FUNCTION(xnn_f16_ibilinear_ukernel__neonfp16arith_c8) DECLARE_F16_IBILINEAR_UKERNEL_FUNCTION(xnn_f16_ibilinear_ukernel__neonfp16arith_c16) #define DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t channels, \ const float** input, \ size_t input_offset, \ const float* weights, \ float* output, \ size_t output_increment); DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__scalar_c1) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__scalar_c2) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__scalar_c4) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__neon_c4) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__neon_c8) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__neonfma_c4) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__neonfma_c8) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__sse_c4) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__sse_c8) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__wasmsimd_c4) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__wasmsimd_c8) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__wasmrelaxedsimd_c4) DECLARE_F32_IBILINEAR_UKERNEL_FUNCTION(xnn_f32_ibilinear_ukernel__wasmrelaxedsimd_c8) #define DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t channels, \ const int8_t** input, \ size_t input_offset, \ const int16_t* weights, \ int8_t* output, \ size_t output_increment); DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__scalar_c1) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__scalar_c2) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__scalar_c4) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__neon_c8) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__neon_c16) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__sse2_c8) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__sse2_c16) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__sse41_c8) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__sse41_c16) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__wasmsimd_dot16x2_c8) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__wasmsimd_dot16x2_c16) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__wasmsimd_mul32_c8) DECLARE_S8_IBILINEAR_UKERNEL_FUNCTION(xnn_s8_ibilinear_ukernel__wasmsimd_mul32_c16) #define DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t channels, \ const uint8_t** input, \ size_t input_offset, \ const int16_t* weights, \ uint8_t* output, \ size_t output_increment); DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__scalar_c1) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__scalar_c2) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__scalar_c4) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__neon_c8) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__neon_c16) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__sse2_c8) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__sse2_c16) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__sse41_c8) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__sse41_c16) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__wasmsimd_dot16x2_c8) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__wasmsimd_dot16x2_c16) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__wasmsimd_mul32_c8) DECLARE_U8_IBILINEAR_UKERNEL_FUNCTION(xnn_u8_ibilinear_ukernel__wasmsimd_mul32_c16) #define DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t channels, \ const float** input, \ size_t input_offset, \ const float* weights, \ float* output, \ size_t input_increment); DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__scalar_p1) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__scalar_p2) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__scalar_p4) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__wasmsimd_p4) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__wasmsimd_p8) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__neon_p4) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__neon_p8) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__neon_p16) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__neonfma_p4) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__neonfma_p8) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__neonfma_p16) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__sse_p4) DECLARE_F32_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f32_ibilinear_chw_ukernel__sse_p8) #define DECLARE_F16_IBILINEAR_CHW_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t channels, \ const void** input, \ size_t input_offset, \ const void* weights, \ void* output, \ size_t input_increment); DECLARE_F16_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f16_ibilinear_chw_ukernel__neonfp16arith_p4) DECLARE_F16_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f16_ibilinear_chw_ukernel__neonfp16arith_p8) DECLARE_F16_IBILINEAR_CHW_UKERNEL_FUNCTION(xnn_f16_ibilinear_chw_ukernel__neonfp16arith_p16) #ifdef __cplusplus } // extern "C" #endif
8,117
46.752941
92
h
XNNPACK
XNNPACK-master/src/xnnpack/indirection.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <stdbool.h> #include <xnnpack.h> #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif XNN_INTERNAL void xnn_indirection_init_conv2d( xnn_operator_t op, size_t output_tile_size, uint32_t log2_element_size); XNN_INTERNAL void xnn_indirection_init_dwconv2d( xnn_operator_t op, size_t step_height, size_t step_width, size_t primary_tile, uint32_t log2_element_size); XNN_INTERNAL void xnn_indirection_init_deconv2d( xnn_operator_t op, size_t output_tile_size, uint32_t log2_element_size); XNN_INTERNAL void xnn_indirection_init_subconv2d( xnn_operator_t op, size_t output_tile_size, uint32_t log2_element_size); XNN_INTERNAL void xnn_indirection_init_maxpool2d( xnn_operator_t op, size_t step_height, size_t step_width, uint32_t log2_element_size); typedef void (*xnn_indirection_init_resize_bilinear2d_hwc_fn)( size_t input_pixel_stride, size_t input_height, size_t input_width, size_t output_height, size_t output_width, const void* input, const void** indirection_buffer, void* packed_weights, bool align_corners, bool tensorflow_legacy); XNN_INTERNAL void xnn_indirection_init_resize_bilinear2d_hwc_f16( size_t input_pixel_stride, size_t input_height, size_t input_width, size_t output_height, size_t output_width, const void* input, const void** indirection_buffer, void* packed_weights, bool align_corners, bool tensorflow_legacy); XNN_INTERNAL void xnn_indirection_init_resize_bilinear2d_hwc_f32( size_t input_pixel_stride, size_t input_height, size_t input_width, size_t output_height, size_t output_width, const void* input, const void** indirection_buffer, float* packed_weights, bool align_corners, bool tensorflow_legacy); XNN_INTERNAL void xnn_indirection_init_resize_bilinear2d_hwc_q11( size_t input_pixel_stride, size_t input_height, size_t input_width, size_t output_height, size_t output_width, const void* input, const void** indirection_buffer, int16_t* packed_weights, bool align_corners, bool tensorflow_legacy); typedef void (*xnn_indirection_init_resize_bilinear2d_chw_fn)( size_t input_pixel_stride, size_t input_height, size_t input_width, size_t output_height, size_t output_width, const void* input, const void** indirection_buffer, void* packed_weights, bool align_corners, bool tensorflow_legacy); XNN_INTERNAL void xnn_indirection_init_resize_bilinear2d_chw_f16( size_t input_pixel_stride, size_t input_height, size_t input_width, size_t output_height, size_t output_width, const void* input, const void** indirection_buffer, void* packed_weights, bool align_corners, bool tensorflow_legacy); XNN_INTERNAL void xnn_indirection_init_resize_bilinear2d_chw_f32( size_t input_pixel_stride, size_t input_height, size_t input_width, size_t output_height, size_t output_width, const void* input, const void** indirection_buffer, float* packed_weights, bool align_corners, bool tensorflow_legacy); XNN_INTERNAL void xnn_indirection_init_unpool2d( xnn_operator_t op, size_t batch_start, uint32_t log2_element_size); typedef void (*xnn_indirection_init_pavgpool2d_fn)( size_t input_height, size_t input_width, size_t output_height, size_t output_width, size_t pooling_height, size_t pooling_width, size_t stride_height, size_t stride_width, size_t padding_top, size_t padding_left, void* pixelwise_buffer); XNN_INTERNAL void xnn_indirection_init_pavgpool2d_f16( size_t input_height, size_t input_width, size_t output_height, size_t output_width, size_t pooling_height, size_t pooling_width, size_t stride_height, size_t stride_width, size_t padding_top, size_t padding_left, uint16_t* pixelwise_buffer); XNN_INTERNAL void xnn_indirection_init_pavgpool2d_f32( size_t input_height, size_t input_width, size_t output_height, size_t output_width, size_t pooling_height, size_t pooling_width, size_t stride_height, size_t stride_width, size_t padding_top, size_t padding_left, float* pixelwise_buffer); #ifdef __cplusplus } // extern "C" #endif
4,445
23.563536
72
h
XNNPACK
XNNPACK-master/src/xnnpack/isa-checks.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <xnnpack/common.h> #include <xnnpack/config.h> #if XNN_ARCH_X86 #define TEST_REQUIRES_X86_SSE \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_SSE #endif #if XNN_ARCH_X86 #define TEST_REQUIRES_X86_SSE2 \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_SSE2 #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_SSSE3 \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_ssse3) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_SSSE3 #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_SSE41 \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_sse4_1) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_SSE41 #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_AVX \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_avx) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_AVX #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_F16C \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_f16c) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_F16C #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_XOP \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_xop) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_XOP #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_FMA3 \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_fma3) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_FMA3 #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_AVX2 \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_avx2) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_AVX2 #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_AVX512F \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_avx512f) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_AVX512F #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_AVX512SKX \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_avx512skx) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_AVX512SKX #endif #if XNN_ARCH_X86 || XNN_ARCH_X86_64 #define TEST_REQUIRES_X86_AVX512VBMI \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_x86_avx512vbmi) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_X86_AVX512VBMI #endif #if XNN_ARCH_ARM #define TEST_REQUIRES_ARM_SIMD32 \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_arm_v6) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_ARM_SIMD32 #endif #if XNN_ARCH_ARM || XNN_ARCH_ARM64 #define TEST_REQUIRES_ARM_FP16_ARITH \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_arm_fp16_arith) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_ARM_FP16_ARITH #endif #if XNN_ARCH_ARM #define TEST_REQUIRES_ARM_NEON \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_arm_neon) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_ARM_NEON #endif #if XNN_ARCH_ARM #define TEST_REQUIRES_ARM_NEON_FP16 \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_arm_neon_fp16) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_ARM_NEON_FP16 #endif #if XNN_ARCH_ARM #define TEST_REQUIRES_ARM_NEON_FMA \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_arm_neon_fma) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_ARM_NEON_FMA #endif #if XNN_ARCH_ARM #define TEST_REQUIRES_ARM_NEON_V8 \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_arm_neon_v8) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_ARM_NEON_V8 #endif #if XNN_ARCH_ARM || XNN_ARCH_ARM64 #define TEST_REQUIRES_ARM_NEON_FP16_ARITH \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_arm_neon_fp16_arith) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_ARM_NEON_FP16_ARITH #endif #if XNN_ARCH_ARM || XNN_ARCH_ARM64 #define TEST_REQUIRES_ARM_NEON_BF16 \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_arm_neon_bf16) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_ARM_NEON_BF16 #endif #if XNN_ARCH_ARM || XNN_ARCH_ARM64 #define TEST_REQUIRES_ARM_NEON_DOT \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_arm_neon_dot) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_ARM_NEON_DOT #endif #if XNN_ARCH_RISCV #define TEST_REQUIRES_RISCV_VECTOR \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_riscv_vector) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_RISCV_VECTOR #endif #if XNN_ARCH_WASMRELAXEDSIMD #define TEST_REQUIRES_WASM_PSHUFB \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_wasm_pshufb) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_WASM_PSHUFB #endif #if XNN_ARCH_WASMRELAXEDSIMD #define TEST_REQUIRES_WASM_SDOT \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_wasm_sdot) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_WASM_SDOT #endif #if XNN_ARCH_WASMRELAXEDSIMD #define TEST_REQUIRES_WASM_BLENDVPS \ do { \ const struct xnn_hardware_config* hardware_config = xnn_init_hardware_config(); \ if (hardware_config == nullptr || !hardware_config->use_wasm_blendvps) { \ GTEST_SKIP(); \ } \ } while (0) #else #define TEST_REQUIRES_WASM_BLENDVPS #endif
8,843
27.165605
87
h
XNNPACK
XNNPACK-master/src/xnnpack/log.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <inttypes.h> #include <stdarg.h> #include <stdlib.h> #include <xnnpack.h> #include <xnnpack/node-type.h> #ifndef XNN_LOG_LEVEL #error "Undefined XNN_LOG_LEVEL" #endif #define XNN_LOG_NONE 0 #define XNN_LOG_FATAL 1 #define XNN_LOG_ERROR 2 #define XNN_LOG_WARNING 3 #define XNN_LOG_INFO 4 #define XNN_LOG_DEBUG 5 #ifdef __cplusplus extern "C" { #endif #if XNN_LOG_LEVEL >= XNN_LOG_DEBUG void xnn_vlog_debug(const char* format, va_list args); #endif #if XNN_LOG_LEVEL >= XNN_LOG_INFO void xnn_vlog_info(const char* format, va_list args); #endif #if XNN_LOG_LEVEL >= XNN_LOG_WARNING void xnn_vlog_warning(const char* format, va_list args); #endif #if XNN_LOG_LEVEL >= XNN_LOG_ERROR void xnn_vlog_error(const char* format, va_list args); #endif #if XNN_LOG_LEVEL >= XNN_LOG_FATAL void xnn_vlog_fatal(const char* format, va_list args); #endif #if XNN_LOG_LEVEL == XNN_LOG_NONE inline static const char* xnn_datatype_to_string(enum xnn_datatype type) { return "Unknown"; } inline static const char* xnn_node_type_to_string(enum xnn_node_type type) { return "Unknown"; } #else const char* xnn_datatype_to_string(enum xnn_datatype type); const char* xnn_node_type_to_string(enum xnn_node_type type); #endif #ifdef __cplusplus } // extern "C" #endif #ifndef XNN_LOG_ARGUMENTS_FORMAT #ifdef __GNUC__ #define XNN_LOG_ARGUMENTS_FORMAT __attribute__((__format__(__printf__, 1, 2))) #else #define XNN_LOG_ARGUMENTS_FORMAT #endif #endif XNN_LOG_ARGUMENTS_FORMAT inline static void xnn_log_debug(const char* format, ...) { #if XNN_LOG_LEVEL >= XNN_LOG_DEBUG va_list args; va_start(args, format); xnn_vlog_debug(format, args); va_end(args); #endif } XNN_LOG_ARGUMENTS_FORMAT inline static void xnn_log_info(const char* format, ...) { #if XNN_LOG_LEVEL >= XNN_LOG_INFO va_list args; va_start(args, format); xnn_vlog_info(format, args); va_end(args); #endif } XNN_LOG_ARGUMENTS_FORMAT inline static void xnn_log_warning(const char* format, ...) { #if XNN_LOG_LEVEL >= XNN_LOG_WARNING va_list args; va_start(args, format); xnn_vlog_warning(format, args); va_end(args); #endif } XNN_LOG_ARGUMENTS_FORMAT inline static void xnn_log_error(const char* format, ...) { #if XNN_LOG_LEVEL >= XNN_LOG_ERROR va_list args; va_start(args, format); xnn_vlog_error(format, args); va_end(args); #endif } XNN_LOG_ARGUMENTS_FORMAT inline static void xnn_log_fatal(const char* format, ...) { #if XNN_LOG_LEVEL >= XNN_LOG_FATAL va_list args; va_start(args, format); xnn_vlog_fatal(format, args); va_end(args); #endif abort(); }
2,923
22.580645
86
h
XNNPACK
XNNPACK-master/src/xnnpack/lut.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_X8_LUT_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t n, \ const uint8_t* x, \ uint8_t* y, \ const uint8_t* t); DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__scalar_x1) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__scalar_x2) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__scalar_x4) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__scalar_x8) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__scalar_x16) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__aarch64_neon_tbx128x4_x16) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__aarch64_neon_tbx128x4_x32) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__aarch64_neon_tbx128x4_x48) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__aarch64_neon_tbx128x4_x64) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__ssse3_x16) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__ssse3_x32) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx_x16) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx_x32) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx_x48) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx_x64) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx2_x32) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx2_x64) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx2_x96) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx2_x128) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx512skx_vpshufb_x64) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx512skx_vpshufb_x128) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx512skx_vpshufb_x192) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx512skx_vpshufb_x256) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx512vbmi_vpermx2b_x64) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx512vbmi_vpermx2b_x128) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx512vbmi_vpermx2b_x192) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__avx512vbmi_vpermx2b_x256) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__wasmpshufb_x16) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__wasmpshufb_x32) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__wasmpshufb_x48) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__wasmpshufb_x64) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__wasmsimd_x16) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__wasmsimd_x32) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__wasmsimd_x48) DECLARE_X8_LUT_UKERNEL_FUNCTION(xnn_x8_lut_ukernel__wasmsimd_x64) #define DECLARE_U8_LUT32NORM_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t n, \ const uint8_t* x, \ const uint32_t* t, \ uint8_t* y); DECLARE_U8_LUT32NORM_UKERNEL_FUNCTION(xnn_u8_lut32norm_ukernel__scalar) #ifdef __cplusplus } // extern "C" #endif
3,483
39.511628
78
h
XNNPACK
XNNPACK-master/src/xnnpack/math.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stdbool.h> #include <stddef.h> #include <stdint.h> #include <assert.h> #ifdef _MSC_VER #include <intrin.h> #include <stdlib.h> // For _rotl. #endif #include <xnnpack/common.h> // stdlib.h from Windows 10 SDK defines min & max macros. // Undefine them before defining the corresponding functions. #ifdef min #undef min #endif #ifdef max #undef max #endif XNN_INLINE static size_t min(size_t a, size_t b) { return XNN_UNPREDICTABLE(b < a) ? b : a; } XNN_INLINE static size_t max(size_t a, size_t b) { return XNN_UNPREDICTABLE(b < a) ? a : b; } XNN_INLINE static size_t doz(size_t a, size_t b) { return XNN_UNPREDICTABLE(b < a) ? a - b : 0; } XNN_INLINE static size_t divide_round_up(size_t n, size_t q) { return XNN_UNPREDICTABLE(n % q == 0) ? n / q : n / q + 1; } XNN_INLINE static size_t round_up(size_t n, size_t q) { return divide_round_up(n, q) * q; } XNN_INLINE static bool is_po2(size_t n) { return (n != 0) && ((n & (n - 1)) == 0); } XNN_INLINE static size_t round_down_po2(size_t n, size_t q) { assert(is_po2(q)); return n & -q; } XNN_INLINE static size_t round_up_po2(size_t n, size_t q) { return round_down_po2(n + q - 1, q); } XNN_INLINE static size_t mod_po2(size_t n, size_t m) { assert(is_po2(m)); return n & (m - 1); } XNN_INLINE static size_t subtract_modulo(size_t a, size_t b, size_t m) { assert(a < m); assert(b < m); return XNN_UNPREDICTABLE(a >= b) ? a - b : a - b + m; } XNN_INLINE static float uint32_as_float(uint32_t i) { union { uint32_t as_uint32; float as_float; } bits = { i }; return bits.as_float; } XNN_INLINE static uint32_t float_as_uint32(float f) { union { float as_float; uint32_t as_uint32; } bits = { f }; return bits.as_uint32; } XNN_INLINE static double uint64_as_double(uint64_t i) { union { uint64_t as_uint64; double as_double; } bits = { i }; return bits.as_double; } XNN_INLINE static uint64_t double_as_uint64(double f) { union { double as_double; uint64_t as_uint64; } bits = { f }; return bits.as_uint64; } XNN_INLINE static uint32_t math_abs_s32(int32_t n) { #if defined(_MSC_VER) return (uint32_t) abs((int) n); #else return XNN_UNPREDICTABLE(n >= 0) ? (uint32_t) n : -(uint32_t) n; #endif } XNN_INLINE static int32_t math_min_s32(int32_t a, int32_t b) { return XNN_UNPREDICTABLE(a < b) ? a : b; } XNN_INLINE static int32_t math_max_s32(int32_t a, int32_t b) { return XNN_UNPREDICTABLE(a > b) ? a : b; } XNN_INLINE static uint32_t math_min_u32(uint32_t a, uint32_t b) { return XNN_UNPREDICTABLE(a < b) ? a : b; } XNN_INLINE static uint32_t math_max_u32(uint32_t a, uint32_t b) { return XNN_UNPREDICTABLE(a > b) ? a : b; } XNN_INLINE static uint32_t math_doz_u32(uint32_t a, uint32_t b) { return XNN_UNPREDICTABLE(a > b) ? a - b : 0; } XNN_INLINE static int64_t math_mulext_s32(int32_t a, int32_t b) { #if defined(_MSC_VER) && defined(_M_IX86) return (int64_t) __emul((int) a, (int) b); #else return (int64_t) a * (int64_t) b; #endif } XNN_INLINE static uint64_t math_mulext_u32(uint32_t a, uint32_t b) { #if defined(_MSC_VER) && defined(_M_IX86) return (uint64_t) __emulu((unsigned int) a, (unsigned int) b); #else return (uint64_t) a * (uint64_t) b; #endif } XNN_INLINE static float math_muladd_f32(float x, float y, float acc) { #if defined(__GNUC__) && defined(__FP_FAST_FMAF) return __builtin_fmaf(x, y, acc); #elif defined(__clang__) && defined(__riscv) return __builtin_fmaf(x, y, acc); #else return x * y + acc; #endif } XNN_INLINE static float math_pmin_f32(float a, float b) { return XNN_UNPREDICTABLE(b < a) ? b : a; } XNN_INLINE static float math_pmax_f32(float a, float b) { return XNN_UNPREDICTABLE(b < a) ? a : b; } XNN_INLINE static double math_pmin_f64(double a, double b) { return XNN_UNPREDICTABLE(b < a) ? b : a; } XNN_INLINE static double math_pmax_f64(double a, double b) { return XNN_UNPREDICTABLE(b < a) ? a : b; } XNN_INLINE static float math_min_f32(float a, float b) { #if defined(__GNUC__) && defined(__ARM_ARCH) && (__ARM_ARCH >= 8) return __builtin_fminf(a, b); #elif defined(__clang__) && defined(__riscv) return __builtin_fminf(a, b); #else return XNN_UNPREDICTABLE(b < a) ? b : a; #endif } XNN_INLINE static float math_max_f32(float a, float b) { #if defined(__GNUC__) && defined(__ARM_ARCH) && (__ARM_ARCH >= 8) return __builtin_fmaxf(a, b); #elif defined(__clang__) && defined(__riscv) return __builtin_fmaxf(a, b); #else return XNN_UNPREDICTABLE(b < a) ? a : b; #endif } XNN_INLINE static double math_min_f64(double a, double b) { #if defined(__GNUC__) && defined(__ARM_ARCH) && (__ARM_ARCH >= 8) return __builtin_fmin(a, b); #elif defined(__clang__) && defined(__riscv) return __builtin_fmin(a, b); #else return XNN_UNPREDICTABLE(b < a) ? b : a; #endif } XNN_INLINE static double math_max_f64(double a, double b) { #if defined(__GNUC__) && defined(__ARM_ARCH) && (__ARM_ARCH >= 8) return __builtin_fmax(a, b); #elif defined(__clang__) && defined(__riscv) return __builtin_fmax(a, b); #else return XNN_UNPREDICTABLE(b < a) ? a : b; #endif } XNN_INLINE static float math_nonsign_mask_f32() { #if defined(__INTEL_COMPILER) // Surprisingly, Intel compiler ignores __builtin_nanf payload return _castu32_f32(0x7FFFFFFF); #elif defined(__GNUC__) return __builtin_nanf("0x7FFFFF"); #else union { uint32_t as_word; float as_float; } f; f.as_word = 0x7FFFFFFF; return f.as_float; #endif } #if defined(__clang__) #if __clang_major__ == 3 && __clang_minor__ >= 7 || __clang_major__ > 3 #define XNN_IGNORE_SHIFT_BASE_UB __attribute__((__no_sanitize__("shift-base"))) #else #define XNN_IGNORE_SHIFT_BASE_UB #endif #elif defined(__GNUC__) #if __GNUC__ >= 8 #define XNN_IGNORE_SHIFT_BASE_UB __attribute__((__no_sanitize__("shift-base"))) #elif __GNUC__ == 4 && __GNUC_MINOR__ >= 9 || __GNUC__ > 4 // 4.9 <= gcc < 8 support ubsan, but doesn't support no_sanitize attribute #define XNN_IGNORE_SHIFT_BASE_UB #ifndef XNN_USE_SHIFT_BASE_UB_WORKAROUND #define XNN_USE_SHIFT_BASE_UB_WORKAROUND 1 #endif #else #define XNN_IGNORE_SHIFT_BASE_UB #endif #else #define XNN_IGNORE_SHIFT_BASE_UB #endif XNN_IGNORE_SHIFT_BASE_UB XNN_INLINE static int32_t math_asr_s32(int32_t x, uint32_t n) { #ifdef XNN_USE_SHIFT_BASE_UB_WORKAROUND #if XNN_ARCH_X86_64 || XNN_ARCH_ARM64 return (int32_t) ((uint64_t) (int64_t) x >> n); #else return x >= 0 ? x >> n : ~(~x >> n); #endif #else return x >> n; #endif } XNN_IGNORE_SHIFT_BASE_UB XNN_INLINE static int64_t math_asr_s64(int64_t x, uint32_t n) { #ifdef XNN_USE_SHIFT_BASE_UB_WORKAROUND return x >= 0 ? x >> n : ~(~x >> n); #else return x >> n; #endif } XNN_INLINE static uint32_t math_clz_u32(uint32_t x) { #if defined(_MSC_VER) && !defined(__clang__) unsigned long index; if XNN_UNPREDICTABLE(_BitScanReverse(&index, (unsigned long) x) != 0) { return (uint32_t) index ^ 31; } else { return 32; } #else if XNN_UNPREDICTABLE(x == 0) { return 32; } else { return (uint32_t) __builtin_clz((unsigned int) x); } #endif } XNN_INLINE static uint32_t math_clz_nonzero_u32(uint32_t x) { assert(x != 0); #if defined(_MSC_VER) && !defined(__clang__) unsigned long index; _BitScanReverse(&index, (unsigned long) x); return (uint32_t) index ^ 31; #else return (uint32_t) __builtin_clz((unsigned int) x); #endif } XNN_INLINE static uint32_t math_ctz_u32(uint32_t x) { #if defined(_MSC_VER) && !defined(__clang__) unsigned long index; _BitScanForward(&index, (unsigned long) x); return (uint32_t) index; #else return (uint32_t) __builtin_ctz((unsigned int) x); #endif } XNN_INLINE static uint32_t math_rotl_u32(uint32_t x, int8_t r) { #if XNN_COMPILER_MSVC return _rotl((unsigned int) x, (int) r); #else return (x << r) | (x >> (32 - r)); #endif } #ifndef __cplusplus XNN_INLINE static uint32_t math_cvt_sat_u32_f64(double x) { #if defined(__GNUC__) && defined(__arm__) uint32_t i; __asm__ ("vcvt.u32.f64 %[i], %P[x]" : [i] "=t" (i) : [x] "w" (x)); return i; #elif defined(__GNUC__) && defined(__aarch64__) uint32_t i; __asm__ ("fcvtnu %w[i], %d[x]" : [i] "=r" (i) : [x] "w" (x)); return i; #elif defined(__GNUC__) && defined(__riscv) uint32_t i; __asm__ ("fcvt.wu.d %[i], %[x], rne" : [i] "=r" (i) : [x] "f" (x)); return i; #elif defined(__clang__) && defined(__wasm__) && defined(__wasm_nontrapping_fptoint__) return __builtin_wasm_trunc_saturate_u_i32_f64(rint(x)); #else x = math_max_f64(x, 0.0); x = math_min_f64(x, 4294967295.0); return (uint32_t) double_as_uint64(x + 0x1.0p+52); #endif } #endif
9,198
24.839888
88
h
XNNPACK
XNNPACK-master/src/xnnpack/maxpool.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #include <xnnpack/microparams.h> #ifdef __cplusplus extern "C" { #endif #define DECLARE_F16_MAXPOOL_MINMAX_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_size, \ size_t channels, \ const void** input, \ size_t input_offset, \ void* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_f16_minmax_params* params); DECLARE_F16_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_f16_maxpool_minmax_ukernel_9p8x__f16c_c8) DECLARE_F16_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_f16_maxpool_minmax_ukernel_9p8x__neonfp16arith_c8) #define DECLARE_F32_MAXPOOL_MINMAX_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_size, \ size_t channels, \ const float** input, \ size_t input_offset, \ float* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_f32_minmax_params* params); DECLARE_F32_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_f32_maxpool_minmax_ukernel_9p8x__neon_c4) DECLARE_F32_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_f32_maxpool_minmax_ukernel_9p8x__scalar_c1) DECLARE_F32_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_f32_maxpool_minmax_ukernel_9p8x__sse_c4) DECLARE_F32_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_f32_maxpool_minmax_ukernel_9p8x__wasm_c1) DECLARE_F32_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_f32_maxpool_minmax_ukernel_9p8x__wasmsimd_arm_c4) DECLARE_F32_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_f32_maxpool_minmax_ukernel_9p8x__wasmsimd_x86_c4) #define DECLARE_U8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_size, \ size_t channels, \ const uint8_t** input, \ size_t input_offset, \ uint8_t* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_u8_minmax_params* params); DECLARE_U8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_u8_maxpool_minmax_ukernel_9p8x__neon_c16) DECLARE_U8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_u8_maxpool_minmax_ukernel_9p8x__scalar_c1) DECLARE_U8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_u8_maxpool_minmax_ukernel_9p8x__sse2_c16) DECLARE_U8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_u8_maxpool_minmax_ukernel_9p8x__wasmsimd_c16) #define DECLARE_S8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(fn_name) \ XNN_INTERNAL void fn_name( \ size_t output_pixels, \ size_t kernel_size, \ size_t channels, \ const int8_t** input, \ size_t input_offset, \ int8_t* output, \ size_t input_increment, \ size_t output_increment, \ const union xnn_s8_minmax_params* params); DECLARE_S8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_s8_maxpool_minmax_ukernel_2p2x__neon_c16) DECLARE_S8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_s8_maxpool_minmax_ukernel_4p3x__neon_c16) DECLARE_S8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_s8_maxpool_minmax_ukernel_9p8x__neon_c16) DECLARE_S8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_s8_maxpool_minmax_ukernel_9p8x__scalar_c1) DECLARE_S8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_s8_maxpool_minmax_ukernel_9p8x__sse2_c16) DECLARE_S8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_s8_maxpool_minmax_ukernel_9p8x__sse41_c16) DECLARE_S8_MAXPOOL_MINMAX_UKERNEL_FUNCTION(xnn_s8_maxpool_minmax_ukernel_9p8x__wasmsimd_c16) #ifdef __cplusplus } // extern "C" #endif
4,891
47.92
98
h
XNNPACK
XNNPACK-master/src/xnnpack/memory-planner.h
// Copyright 2020 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stdbool.h> #include <stddef.h> #include <stdint.h> #include <xnnpack.h> #include <xnnpack/common.h> #include <xnnpack/allocator.h> #ifdef __cplusplus extern "C" { #endif struct xnn_usage_record { // The index (to xnn_runtime_t->opdata) of the first xnn_node that uses this xnn_value. uint32_t first_node; // The index of the last xnn_node that uses this xnn_value. uint32_t last_node; // Note that 'tensor_size' includes the padding of XNN_EXTRA_BYTES. size_t tensor_size; // The memory offset of this xnn_value from the beginning of a memory buffer. size_t alloc_offset; // If an operation is performed in place, the alloc_offset of the output tensor is the same as the alloc_offset of the // input tensor. The id of the input tensor is recorded in this field. This is XNN_INVALID_VALUE_ID if it does not // reuse any tensor. uint32_t reuse_value_id; // This usage record is not tied to an actual value, but a temporary associated with an opdata, like a dynamic fully // connected operation. We need the opdata's id to lookup and intialize opdata's pointers. uint32_t opdata_id; }; // Track the memory allocation in a memory arena for a runtime. struct xnn_value_allocation_tracker { size_t mem_arena_size; // Representing the lifecycle of xnn_values in the 'runtime', and the array size is 'runtime->num_values + // runtime->num_ops'. struct xnn_usage_record* usage; // The range of value ids (i.e. the index to runtime->values) whose memory might need to be allocated. size_t min_value_id; size_t max_value_id; }; // Initialize the memory allocation tracker for xnn_values. XNN_INTERNAL void xnn_init_value_allocation_tracker(struct xnn_value_allocation_tracker* tracker, const struct xnn_runtime* runtime); inline static void xnn_release_value_allocation_tracker(struct xnn_value_allocation_tracker* tracker) { xnn_release_memory(tracker->usage); } // Add a to-be-allocated xnn_value (referred by 'value_id') of size 'tensor_size' to the allocation tracker. // Note: this function assumes 'value_id's added in increasing order for simplicity as it's called inside a loop // iterating over 'runtime->values'. XNN_INTERNAL void xnn_add_value_allocation_tracker(struct xnn_value_allocation_tracker* tracker, uint32_t value_id, size_t tensor_size); // Add a value to represent operator workspace. This is a temporary buffer that is only used during the invocation of // operator. XNN_INTERNAL void xnn_add_operator_workspace_allocation_tracker( struct xnn_value_allocation_tracker* tracker, uint32_t operator_workspace_value_id, size_t tensor_size, uint32_t opdata_id); // Mark value_id as reusing the memory that is allocated to another reuse_value_id. No memory is then // allocated to value_id. The usage record of reuse_value_id needs to be expanded to include al consumers of value_id, // indicated by new_last_node. XNN_INTERNAL void xnn_mark_tensor_as_reuse( struct xnn_value_allocation_tracker* tracker, uint32_t value_id, uint32_t reuse_value_id, uint32_t new_last_node); // Plan the exact the memory allocation for intermediate tensors according to the xnn_value allocation tracker. XNN_INTERNAL void xnn_plan_value_allocation_tracker(struct xnn_value_allocation_tracker* tracker); #ifdef __cplusplus } // extern "C" #endif
3,600
40.872093
120
h
XNNPACK
XNNPACK-master/src/xnnpack/microkernel-type.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. // // Auto-generated file. Do not edit! // Specification: src/enums/microkernel-type.yaml // Generator: tools/generate-enum.py #pragma once #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif enum xnn_microkernel_type { xnn_microkernel_type_default = 0, xnn_microkernel_type_average_pooling, xnn_microkernel_type_conv2d_hwc2chw, xnn_microkernel_type_dwconv, xnn_microkernel_type_gemm, xnn_microkernel_type_global_average_pooling, xnn_microkernel_type_igemm, xnn_microkernel_type_mean, xnn_microkernel_type_pixelwise_average_pooling, xnn_microkernel_type_spmm, xnn_microkernel_type_subconv2d, xnn_microkernel_type_transpose, xnn_microkernel_type_vmulcaddc, }; XNN_INTERNAL const char* xnn_microkernel_type_to_string(enum xnn_microkernel_type microkernel_type); #ifdef __cplusplus } // extern "C" #endif
1,019
24.5
100
h
XNNPACK
XNNPACK-master/src/xnnpack/microkernel-utils.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif // The total tile size needed to cover kernel_size. XNN_INTERNAL size_t xnn_dwconv_multipass_tile_size( size_t kernel_size, size_t first_pass_tile, size_t middle_pass_tile, size_t last_pass_tile); // The total count of weights (in bytes) needed for multipass dwconv. size_t xnn_dwconv_multipass_weights_size( size_t tile_size, size_t channels, size_t channel_tile, size_t channel_subtile, size_t channel_round, size_t bias_element_size, size_t log2_filter_element_size, size_t extra_weights_byte); // Calculate the number of bytes read. size_t xnn_dwconv_multipass_bytes_read( size_t kernel_size, size_t first_pass_tile, size_t middle_pass_tile, size_t last_pass_tile, size_t channels, size_t channel_tile, size_t channel_subtile, size_t channel_round, size_t log2_input_size, size_t log2_filter_size, size_t bias_element_size, size_t log2_accumulator_size); // Calculate the number of bytes written. size_t xnn_dwconv_multipass_bytes_written( size_t kernel_size, size_t first_pass_tile, size_t middle_pass_tile, size_t last_pass_tile, size_t channels, size_t channel_round, size_t log2_accumulator_size, size_t log2_output_size); #ifdef __cplusplus } #endif
1,496
22.761905
72
h
XNNPACK
XNNPACK-master/src/xnnpack/models.h
// Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <xnnpack.h> #include <memory> #include <vector> namespace models { typedef std::vector<std::unique_ptr<xnn_operator, decltype(&xnn_delete_operator)>> ExecutionPlan; typedef ExecutionPlan (*ExecutionPlanFactory)(pthreadpool_t threadpool); ExecutionPlan FP32MobileNetV1(pthreadpool_t threadpool); ExecutionPlan FP32MobileNetV2(pthreadpool_t threadpool); ExecutionPlan FP32MobileNetV3Large(pthreadpool_t threadpool); ExecutionPlan FP32MobileNetV3Small(pthreadpool_t threadpool); ExecutionPlan FP32MobileNetV1Jit(pthreadpool_t threadpool); ExecutionPlan FP32MobileNetV2Jit(pthreadpool_t threadpool); ExecutionPlan FP32MobileNetV3LargeJit(pthreadpool_t threadpool); ExecutionPlan FP32MobileNetV3SmallJit(pthreadpool_t threadpool); ExecutionPlan FP32MobileNetV3SmallFused(pthreadpool_t threadpool); ExecutionPlan FP32SparseMobileNetV1(float sparsity, pthreadpool_t threadpool); ExecutionPlan FP32SparseMobileNetV2(float sparsity, pthreadpool_t threadpool); ExecutionPlan FP32SparseMobileNetV3Large(float sparsity, pthreadpool_t threadpool); ExecutionPlan FP32SparseMobileNetV3Small(float sparsity, pthreadpool_t threadpool); ExecutionPlan FP16MobileNetV1(pthreadpool_t threadpool); ExecutionPlan FP16MobileNetV2(pthreadpool_t threadpool); ExecutionPlan FP16MobileNetV3Large(pthreadpool_t threadpool); ExecutionPlan FP16MobileNetV3Small(pthreadpool_t threadpool); ExecutionPlan FP16SparseMobileNetV1(float sparsity, pthreadpool_t threadpool); ExecutionPlan FP16SparseMobileNetV2(float sparsity, pthreadpool_t threadpool); ExecutionPlan FP16SparseMobileNetV3Large(float sparsity, pthreadpool_t threadpool); ExecutionPlan FP16SparseMobileNetV3Small(float sparsity, pthreadpool_t threadpool); ExecutionPlan QC8MobileNetV1(pthreadpool_t threadpool); ExecutionPlan QC8MobileNetV2(pthreadpool_t threadpool); ExecutionPlan QS8MobileNetV1(pthreadpool_t threadpool); ExecutionPlan QS8MobileNetV2(pthreadpool_t threadpool); ExecutionPlan QU8MobileNetV1(pthreadpool_t threadpool); ExecutionPlan QU8MobileNetV2(pthreadpool_t threadpool); } // namespace models
2,242
40.537037
97
h
XNNPACK
XNNPACK-master/src/xnnpack/mutex.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <xnnpack.h> #include <xnnpack/common.h> #if XNN_PLATFORM_WINDOWS #ifndef WIN32_LEAN_AND_MEAN #define WIN32_LEAN_AND_MEAN #endif #include <windows.h> #elif XNN_PLATFORM_MACOS || XNN_PLATFORM_IOS #include <dispatch/dispatch.h> #else #include <pthread.h> #endif #ifdef __cplusplus extern "C" { #endif struct xnn_mutex { #if XNN_PLATFORM_WINDOWS HANDLE handle; #elif XNN_PLATFORM_MACOS || XNN_PLATFORM_IOS dispatch_semaphore_t semaphore; #elif XNN_PLATFORM_WEB && !defined(__EMSCRIPTEN_PTHREADS__) char _; // Dummy member variable to comply with the C standard #else pthread_mutex_t mutex; #endif }; enum xnn_status xnn_mutex_init(struct xnn_mutex* mutex); enum xnn_status xnn_mutex_lock(struct xnn_mutex* mutex); enum xnn_status xnn_mutex_unlock(struct xnn_mutex* mutex); enum xnn_status xnn_mutex_destroy(struct xnn_mutex* mutex); #ifdef __cplusplus } // extern "C" #endif
1,072
22.326087
72
h
XNNPACK
XNNPACK-master/src/xnnpack/node-type.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once enum xnn_node_type { xnn_node_type_invalid = 0, xnn_node_type_abs, xnn_node_type_add2, xnn_node_type_argmax_pooling_2d, xnn_node_type_average_pooling_2d, xnn_node_type_bankers_rounding, xnn_node_type_batch_matrix_multiply, xnn_node_type_ceiling, xnn_node_type_clamp, xnn_node_type_concatenate2, xnn_node_type_concatenate3, xnn_node_type_concatenate4, xnn_node_type_convert, xnn_node_type_convolution_2d, xnn_node_type_copy, xnn_node_type_deconvolution_2d, xnn_node_type_depth_to_space, xnn_node_type_depthwise_convolution_2d, xnn_node_type_divide, xnn_node_type_elu, xnn_node_type_even_split2, xnn_node_type_even_split3, xnn_node_type_even_split4, xnn_node_type_floor, xnn_node_type_fully_connected, xnn_node_type_fully_connected_sparse, xnn_node_type_global_average_pooling_1d, xnn_node_type_global_average_pooling_2d, xnn_node_type_global_sum_pooling_1d, xnn_node_type_global_sum_pooling_2d, xnn_node_type_hardswish, xnn_node_type_leaky_relu, xnn_node_type_max_pooling_2d, xnn_node_type_maximum2, xnn_node_type_minimum2, xnn_node_type_multiply2, xnn_node_type_negate, xnn_node_type_prelu, xnn_node_type_rope, xnn_node_type_sigmoid, xnn_node_type_softmax, xnn_node_type_space_to_depth_2d, xnn_node_type_square, xnn_node_type_square_root, xnn_node_type_squared_difference, xnn_node_type_static_constant_pad, xnn_node_type_static_mean, xnn_node_type_static_reshape, xnn_node_type_static_resize_bilinear_2d, xnn_node_type_static_slice, xnn_node_type_static_transpose, xnn_node_type_subtract, xnn_node_type_tanh, xnn_node_type_unpooling_2d, };
1,828
27.578125
72
h
XNNPACK
XNNPACK-master/src/xnnpack/normalization.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <xnnpack.h> #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif // Calculates normalized offsets, input_shape, and output_shape. // Each value in offsets must be less than the corresponding dimension of input_shape. // Each value in sizes must be > 0 and less than or equals to the corresponding dimension of input_shape. // This function merges dimensions dimensions that are full slices into the outermost dimension possible. // E.g. Given input shape { 4, 5, 3 }, with offsets { 0, 2, 0 }, and sizes { 4, 1, 3 }, the innermost dimension is a // full slice, and so can be merged with its outer dimension, to give normalized input shape of { 4, 15 }, // output shape { 4, 3 } with offsets { 0, 6 }. void xnn_normalize_slice( size_t num_dims, const size_t offsets[XNN_MIN_ELEMENTS(1)], const size_t sizes[XNN_MIN_ELEMENTS(1)], const size_t input_shape[XNN_MIN_ELEMENTS(1)], size_t normalized_offsets[XNN_MIN_ELEMENTS(XNN_MAX_TENSOR_DIMS)], size_t normalized_input_shape[XNN_MIN_ELEMENTS(XNN_MAX_TENSOR_DIMS)], size_t normalized_output_shape[XNN_MIN_ELEMENTS(XNN_MAX_TENSOR_DIMS)], size_t* num_normalized_dims); void xnn_normalize_transpose_permutation( size_t num_dims, size_t element_size, const size_t* perm, const size_t* shape, const size_t* input_stride, const size_t* output_stride, size_t* normalized_num_dims, size_t* normalized_element_size, size_t* normalized_perm, size_t* normalized_shape, size_t* normalized_input_stride, size_t* normalized_output_stride); void xnn_normalize_reduction( size_t* num_reduction_axes_ptr, size_t* reduction_axes, size_t* num_input_dims_ptr, size_t* input_dims); #ifdef __cplusplus } // extern "C" #endif
1,973
33.631579
116
h
XNNPACK
XNNPACK-master/src/xnnpack/operator-type.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. // // Auto-generated file. Do not edit! // Specification: src/enums/operator-type.yaml // Generator: tools/generate-enum.py #pragma once #include <xnnpack/common.h> #ifdef __cplusplus extern "C" { #endif enum xnn_operator_type { xnn_operator_type_invalid = 0, xnn_operator_type_abs_nc_f16, xnn_operator_type_abs_nc_f32, xnn_operator_type_add_nd_f16, xnn_operator_type_add_nd_f32, xnn_operator_type_add_nd_qs8, xnn_operator_type_add_nd_qu8, xnn_operator_type_argmax_pooling_nhwc_f32, xnn_operator_type_average_pooling_nhwc_f16, xnn_operator_type_average_pooling_nhwc_f32, xnn_operator_type_average_pooling_nhwc_qu8, xnn_operator_type_bankers_rounding_nc_f16, xnn_operator_type_bankers_rounding_nc_f32, xnn_operator_type_batch_matrix_multiply_nc_f32, xnn_operator_type_ceiling_nc_f16, xnn_operator_type_ceiling_nc_f32, xnn_operator_type_channel_shuffle_nc_x8, xnn_operator_type_channel_shuffle_nc_x32, xnn_operator_type_clamp_nc_f16, xnn_operator_type_clamp_nc_f32, xnn_operator_type_clamp_nc_s8, xnn_operator_type_clamp_nc_u8, xnn_operator_type_constant_pad_nd_x8, xnn_operator_type_constant_pad_nd_x16, xnn_operator_type_constant_pad_nd_x32, xnn_operator_type_convert_nc_f16_f32, xnn_operator_type_convert_nc_f32_f16, xnn_operator_type_convert_nc_f32_qd8, xnn_operator_type_convert_nc_f32_qs8, xnn_operator_type_convert_nc_f32_qu8, xnn_operator_type_convert_nc_qs8, xnn_operator_type_convert_nc_qs8_f32, xnn_operator_type_convert_nc_qs16_qs8, xnn_operator_type_convert_nc_qu8, xnn_operator_type_convert_nc_qu8_f32, xnn_operator_type_convolution_nchw_f16, xnn_operator_type_convolution_nchw_f32, xnn_operator_type_convolution_nhwc_f16, xnn_operator_type_convolution_nhwc_f32, xnn_operator_type_convolution_nhwc_qc8, xnn_operator_type_convolution_nhwc_qs8, xnn_operator_type_convolution_nhwc_qu8, xnn_operator_type_copy_nc_x8, xnn_operator_type_copy_nc_x16, xnn_operator_type_copy_nc_x32, xnn_operator_type_deconvolution_nhwc_f16, xnn_operator_type_deconvolution_nhwc_f32, xnn_operator_type_deconvolution_nhwc_qs8, xnn_operator_type_deconvolution_nhwc_qu8, xnn_operator_type_depth_to_space_nchw2nhwc_x16, xnn_operator_type_depth_to_space_nchw2nhwc_x32, xnn_operator_type_depth_to_space_nhwc_x8, xnn_operator_type_depth_to_space_nhwc_x16, xnn_operator_type_depth_to_space_nhwc_x32, xnn_operator_type_divide_nd_f16, xnn_operator_type_divide_nd_f32, xnn_operator_type_dynamic_fully_connected_nc_f16, xnn_operator_type_dynamic_fully_connected_nc_f32, xnn_operator_type_elu_nc_f16, xnn_operator_type_elu_nc_f32, xnn_operator_type_elu_nc_qs8, xnn_operator_type_floor_nc_f16, xnn_operator_type_floor_nc_f32, xnn_operator_type_fully_connected_nc_f16, xnn_operator_type_fully_connected_nc_f32, xnn_operator_type_fully_connected_nc_f32_qc8w, xnn_operator_type_fully_connected_nc_qs8, xnn_operator_type_fully_connected_nc_qu8, xnn_operator_type_global_average_pooling_ncw_f16, xnn_operator_type_global_average_pooling_ncw_f32, xnn_operator_type_global_average_pooling_nwc_f16, xnn_operator_type_global_average_pooling_nwc_f32, xnn_operator_type_global_average_pooling_nwc_qs8, xnn_operator_type_global_average_pooling_nwc_qu8, xnn_operator_type_global_sum_pooling_nwc_f16, xnn_operator_type_global_sum_pooling_nwc_f32, xnn_operator_type_hardswish_nc_f16, xnn_operator_type_hardswish_nc_f32, xnn_operator_type_leaky_relu_nc_f16, xnn_operator_type_leaky_relu_nc_f32, xnn_operator_type_leaky_relu_nc_qs8, xnn_operator_type_leaky_relu_nc_qu8, xnn_operator_type_max_pooling_nhwc_f16, xnn_operator_type_max_pooling_nhwc_f32, xnn_operator_type_max_pooling_nhwc_s8, xnn_operator_type_max_pooling_nhwc_u8, xnn_operator_type_maximum_nd_f16, xnn_operator_type_maximum_nd_f32, xnn_operator_type_mean_nd_f16, xnn_operator_type_mean_nd_f32, xnn_operator_type_minimum_nd_f16, xnn_operator_type_minimum_nd_f32, xnn_operator_type_multiply_nd_f16, xnn_operator_type_multiply_nd_f32, xnn_operator_type_multiply_nd_qs8, xnn_operator_type_multiply_nd_qu8, xnn_operator_type_negate_nc_f16, xnn_operator_type_negate_nc_f32, xnn_operator_type_prelu_nc_f16, xnn_operator_type_prelu_nc_f32, xnn_operator_type_resize_bilinear_nchw_f16, xnn_operator_type_resize_bilinear_nchw_f32, xnn_operator_type_resize_bilinear_nhwc_f16, xnn_operator_type_resize_bilinear_nhwc_f32, xnn_operator_type_resize_bilinear_nhwc_s8, xnn_operator_type_resize_bilinear_nhwc_u8, xnn_operator_type_rope_nthc_f32, xnn_operator_type_sigmoid_nc_f16, xnn_operator_type_sigmoid_nc_f32, xnn_operator_type_sigmoid_nc_qs8, xnn_operator_type_sigmoid_nc_qu8, xnn_operator_type_slice_nd_x8, xnn_operator_type_slice_nd_x16, xnn_operator_type_slice_nd_x32, xnn_operator_type_softmax_nc_f16, xnn_operator_type_softmax_nc_f32, xnn_operator_type_softmax_nc_qu8, xnn_operator_type_space_to_depth_nhwc_x8, xnn_operator_type_space_to_depth_nhwc_x16, xnn_operator_type_space_to_depth_nhwc_x32, xnn_operator_type_square_nc_f16, xnn_operator_type_square_nc_f32, xnn_operator_type_square_root_nc_f16, xnn_operator_type_square_root_nc_f32, xnn_operator_type_squared_difference_nd_f16, xnn_operator_type_squared_difference_nd_f32, xnn_operator_type_subtract_nd_f16, xnn_operator_type_subtract_nd_f32, xnn_operator_type_subtract_nd_qs8, xnn_operator_type_subtract_nd_qu8, xnn_operator_type_tanh_nc_f16, xnn_operator_type_tanh_nc_f32, xnn_operator_type_tanh_nc_qs8, xnn_operator_type_tanh_nc_qu8, xnn_operator_type_transpose_nd_x8, xnn_operator_type_transpose_nd_x16, xnn_operator_type_transpose_nd_x32, xnn_operator_type_truncation_nc_f16, xnn_operator_type_truncation_nc_f32, xnn_operator_type_unpooling_nhwc_x32, }; XNN_INTERNAL const char* xnn_operator_type_to_string(enum xnn_operator_type operator_type); #ifdef __cplusplus } // extern "C" #endif
6,116
35.628743
91
h
XNNPACK
XNNPACK-master/src/xnnpack/operator-utils.h
// Copyright 2022 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <xnnpack/common.h> #include <xnnpack/operator.h> #include <xnnpack/params.h> #if XNN_PLATFORM_JIT // Generates code for all mr values up to max_mr. // Offsets of all generated code will be kept in generated_code_offset. XNN_INTERNAL void xnn_generate_gemms_up_to_max_mr( size_t max_mr, struct gemm_codegens generators, const struct jit_gemm_params *jit_gemm_params, size_t group_output_channels, size_t nr, size_t group_input_channels_in_bytes, xnn_operator_t convolution_op); XNN_INTERNAL void xnn_generate_igemms_up_to_max_mr( size_t max_mr, struct gemm_codegens generators, const struct jit_gemm_params *jit_gemm_params, size_t group_output_channels, size_t nr, size_t group_input_channels_in_bytes, size_t kernel_size, xnn_operator_t convolution_op); // Overwrite function pointer to GEMM microkernels with generated code if available. XNN_INTERNAL void xnn_overwrite_gemm_cases_with_generated_code( xnn_operator_t convolution_op, struct xnn_hmp_gemm_ukernel *gemm_cases, size_t mr); // Overwrite function pointer to IGEMM microkernels with generated code if available. XNN_INTERNAL void xnn_overwrite_igemm_cases_with_generated_code( xnn_operator_t convolution_op, struct xnn_hmp_igemm_ukernel *igemm_cases, size_t mr); XNN_INTERNAL void xnn_generate_vunary_ukernel( const struct xnn_unary_elementwise_config* config, xnn_operator_t op); #endif // XNN_PLATFORM_JIT static inline void* packed_weights(struct xnn_operator* op) { if (op->weights_cache == NULL) { return op->packed_weights.pointer; } else { return (void*) ((uintptr_t) op->weights_cache->cache.weights.start + op->packed_weights.offset); } } static inline bool use_weights_cache(struct xnn_operator* op) { return op->weights_cache != NULL; } // Get a pointer to a region to pack weights into. If weights cache is available, use it, returning to a pointer to the // cache's buffer, otherwise, allocate and return a pointer to a new region. Returns NULL on error. XNN_INTERNAL void* xnn_get_pointer_to_write_weights( xnn_operator_t op, size_t aligned_weights_size, int padding_byte); #ifdef __cplusplus extern "C" { #endif XNN_INTERNAL size_t xnn_compute_convolution_output_dimension( size_t padded_input_dimension, size_t kernel_dimension, size_t dilation_dimension, size_t subsampling_dimension); XNN_INTERNAL size_t xnn_compute_deconvolution_output_dimension( size_t input_dimension, size_t output_padding_dimension, size_t adjustment_dimension, size_t kernel_dimension, size_t dilation_dimension, size_t stride_dimension); XNN_INTERNAL size_t xnn_compute_unpooling_output_dimension( size_t input_dimension, size_t input_padding_dimension, size_t kernel_dimension); XNN_INTERNAL uint32_t xnn_get_heuristic_mr_gemm( size_t batch_size, uint32_t max_mr, uint32_t nr, struct xnn_hmp_gemm_ukernel *gemm_cases, bool code_cache_available); XNN_INTERNAL uint32_t xnn_get_heuristic_mr_igemm( size_t batch_size, uint32_t max_mr, uint32_t nr, struct xnn_hmp_igemm_ukernel *igemm_cases, bool code_cache_available); #ifdef __cplusplus } #endif
3,370
29.645455
119
h
XNNPACK
XNNPACK-master/src/xnnpack/operator.h
// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include <stddef.h> #include <stdint.h> #include <pthreadpool.h> #include <xnnpack/allocator.h> #include <xnnpack/cache.h> #include <xnnpack/compute.h> #include <xnnpack/config.h> #include <xnnpack/microkernel-type.h> #include <xnnpack/operator-type.h> #include <xnnpack/params.h> // Maximum number of pthreadpool parallelization invocations per operator. #define XNN_MAX_COMPUTE_INVOCATIONS 2 struct xnn_ukernel_conv2d { union { xnn_conv_hwc2chw_ukernel_fn hwc2chw_fn; xnn_conv_hwc_ukernel_fn hwc_fn; }; uint8_t output_height_tile; uint8_t output_channel_tile; }; struct xnn_ukernel_dwconv { union { xnn_dwconv_unipass_ukernel_fn unipass_fn; xnn_dwconv_multipass_ukernel_fn multipass_fn; }; uint8_t primary_tile; uint8_t middle_tile; uint8_t last_tile; // For unipass, tile_size == primary_tile, otherwise it is calculated based on // how many pass the middle_tile runs. uint8_t tile_size; }; // Direct 2D Depthwise Convolution struct xnn_ukernel_dwconv2d { union { xnn_dwconv2d_chw_ukernel_fn chw_fn; }; xnn_update_chw_params_fn update_params; uint8_t output_width_tile; }; struct xnn_ukernel_gemm { struct xnn_hmp_gemm_ukernel gemm_cases[XNN_MAX_MR]; xnn_packw_gemm_goi_ukernel_fn packw_gemm_goi; uint8_t mr; uint8_t nr; uint8_t kr; uint8_t sr; }; struct xnn_ukernel_igemm { struct xnn_hmp_igemm_ukernel igemm_cases[XNN_MAX_MR]; struct xnn_hmp_gemm_ukernel gemm_cases[XNN_MAX_MR]; uint8_t mr; uint8_t nr; uint8_t kr; uint8_t sr; }; struct xnn_ukernel_spmm { xnn_spmm_ukernel_fn function; uint8_t mr; }; struct xnn_ukernel_vmulcaddc { xnn_vmulcaddc_ukernel_fn function; uint8_t mr; }; struct xnn_ukernel_vbinary { xnn_vbinary_ukernel_fn op_fn; xnn_vbinary_ukernel_fn opc_fn; xnn_vbinary_ukernel_fn ropc_fn; }; struct xnn_ukernel_vunary { xnn_vunary_ukernel_fn function; }; struct xnn_ukernel { enum xnn_microkernel_type type; // Used by subconv2d whether it is a GEMM or IGEMM. enum xnn_microkernel_type subtype; union { struct xnn_ukernel_conv2d conv2d; struct xnn_ukernel_dwconv dwconv; struct xnn_ukernel_dwconv2d dwconv2d; struct { struct xnn_ukernel_gemm gemm; struct xnn_ukernel_gemm gemm2; }; struct xnn_ukernel_igemm igemm; struct xnn_ukernel_spmm spmm; struct xnn_ukernel_vmulcaddc vmulcaddc; struct xnn_ukernel_vbinary vbinary; struct xnn_ukernel_vunary vunary; }; }; // Valid state transitions: // - xnn_run_state_invalid -> xnn_run_state_skip // - xnn_run_state_invalid -> xnn_run_state_ready // - xnn_run_state_invalid -> xnn_run_state_needs_setup -> xnn_run_state_ready enum xnn_run_state { // When an operator is first created, it starts off in invalid state, it needs to be setup, or reshape + setup. xnn_run_state_invalid = 0, // Operator is ready to be run. xnn_run_state_ready, // Operator doesn't need to be run. xnn_run_state_skip, // Operator has been reshaped, but not setup yet, pointers are not set. xnn_run_state_needs_setup, }; struct subconvolution_params { void* weights; size_t w_stride; const void** indirection_buffer; void* output; size_t slice_width; size_t slice_height; size_t indirection_y_stride; size_t indirection_x_stride; // scaled_kernel_size := kernel_size * mr * sizeof(void*). size_t scaled_kernel_size; }; struct xnn_operator { size_t batch_size; uint32_t padding_top; uint32_t padding_right; uint32_t padding_bottom; uint32_t padding_left; uint32_t kernel_height; uint32_t kernel_width; uint32_t stride_height; uint32_t stride_width; uint32_t dilation_height; uint32_t dilation_width; uint32_t groups; size_t group_channels; size_t group_input_channels; size_t group_output_channels; size_t channels; size_t max_sequence_size; uint32_t pad_value; size_t input_height; size_t input_width; size_t input_pixel_stride; const void* input; const void* input2; const void** indirection_buffer; size_t output_height; size_t output_width; size_t output_pixel_stride; void* output; union { // Pointer to allocated packed weights. Use this if weights_cache is NULL. void* pointer; // Offset into the weights cache where the packed weights are. Only valid if weights_cache is not NULL. size_t offset; } packed_weights; // Total number of non-zero kernel elements when weights use sparse representation. size_t num_nonzero_values; // Total number of non-zero kernel blocks when weights use sparse representation. size_t num_nonzero_blocks; // Total number of output channel blocks when weights use sparse representation. size_t num_output_channel_blocks; // Input channel corresponding to the first non-zero kernel element. size_t first_input_channel; float input_scale; float output_scale; int32_t input_zero_point; size_t valid_batch_size; size_t last_input_height; size_t last_input_width; const void* last_input; size_t last_output_height; size_t last_output_width; void* last_output; uint32_t last_mr; uint32_t block_size; void* zero_buffer; void* lookup_table; void* pixelwise_buffer; struct subconvolution_params* subconvolution_buffer; uint32_t flags; union { union xnn_f16_abs_params f16_abs; union xnn_f16_f32_cvt_params f16_f32_cvt; union xnn_f16_hswish_params f16_hswish; union xnn_f16_elu_params f16_elu; union xnn_f16_lrelu_params f16_lrelu; union xnn_f16_neg_params f16_neg; union xnn_f16_sigmoid_params f16_sigmoid; union xnn_f16_tanh_params f16_tanh; union xnn_f32_abs_params f32_abs; union xnn_f32_default_params f32_default; union xnn_f32_elu_params f32_elu; union xnn_f32_lrelu_params f32_lrelu; union xnn_f32_neg_params f32_neg; union xnn_f32_rnd_params f32_rnd; union xnn_f32_sigmoid_params f32_sigmoid; union xnn_f32_sqrt_params f32_sqrt; union xnn_f32_tanh_params f32_tanh; // Parameters for Global Average Pooling in CHW layout union xnn_f16_gavgpool_params f16_gavgpool; union xnn_f32_gavgpool_params f32_gavgpool; union xnn_f32_hswish_params f32_hswish; // Pixelwise Average Pooling normally use f16_minmax_params, but also initialize // f16_scaleminmax_params in case it needs to switch to Global Average Pooling operation. struct { union xnn_f16_minmax_params f16_minmax; union xnn_f16_scaleminmax_params f16_scaleminmax; }; // Mean can use either f16_f32acc_scale, or f16_scale_minmax struct { union xnn_f16_f32acc_scale_params f16_f32acc_scale; union xnn_f16_scaleminmax_params f16_scale_minmax; }; // Pixelwise Average Pooling normally use f32_minmax_params, but also initialize // f32_scaleminmax_params in case it needs to switch to Global Average Pooling operation. struct { union xnn_f32_minmax_params f32_minmax; union xnn_f32_scaleminmax_params f32_scaleminmax; }; // Mean can use either f32_scale, or f32_scale_minmax struct { union xnn_f32_scale_params f32_scale; union xnn_f32_scaleminmax_params f32_scale_minmax; }; union xnn_f16_chw_params f16_chw; union xnn_f32_chw_params f32_chw; union xnn_f32_f16_cvt_params f32_f16_cvt; union xnn_f32_qs8_cvt_params f32_qs8_cvt; union xnn_f32_qu8_cvt_params f32_qu8_cvt; union xnn_qs8_cvt_params qs8_cvt; union xnn_qs8_f32_cvt_params qs8_f32_cvt; union xnn_qs16_qs8_cvt_params qs16_qs8_cvt; union xnn_qu8_cvt_params qu8_cvt; union xnn_qu8_f32_cvt_params qu8_f32_cvt; union xnn_qs8_conv_minmax_params qs8_conv_minmax; // Average Pooling normally use qs8_avgpool_params, but also initialize qs8_gavgpool_params in case it needs to switch // to Global Average Pooling operation. struct { union xnn_qs8_avgpool_minmax_params qs8_avgpool; union xnn_qs8_avgpool_minmax_params qs8_gavgpool; }; // Quantized Add parameters are sensitive to order of inputs, so we initialize an extra copy with the reversed order. struct { union xnn_qs8_add_minmax_params qs8_add; union xnn_qs8_add_minmax_params qs8_radd; }; struct { union xnn_qs8_mul_minmax_params qs8_mul; union xnn_qs8_mul_minmax_params qs8_rmul; }; struct { union xnn_qu8_add_minmax_params qu8_add; union xnn_qu8_add_minmax_params qu8_radd; }; struct { union xnn_qu8_mul_minmax_params qu8_mul; union xnn_qu8_mul_minmax_params qu8_rmul; }; union xnn_qu8_conv_minmax_params qu8_conv_minmax; // Average Pooling normally use qu8_avgpool_params, but also initialize qu8_gavgpool_params in case it needs to switch // to Global Average Pooling operation. struct { union xnn_qu8_avgpool_minmax_params qu8_avgpool; union xnn_qu8_avgpool_minmax_params qu8_gavgpool; }; union xnn_qs8_hswish_params qs8_hswish; union xnn_qu8_hswish_params qu8_hswish; union xnn_qs8_lrelu_params qs8_lrelu; union xnn_qu8_lrelu_params qu8_lrelu; union xnn_s8_minmax_params s8_minmax; union xnn_u8_minmax_params u8_minmax; } params; // Second set of params. Operators like Dynamic Fully Connected only decides on the specific config to use during // reshape, so it needs to keep two sets of params around. Configs can have different initialization functions. union { union xnn_f32_minmax_params f32_minmax; } params2; size_t num_post_operation_params; void* post_operation_params; enum xnn_operator_type type; struct xnn_ukernel ukernel; union { const struct xnn_argmaxpool_config* argmaxpool_config; struct { const struct xnn_avgpool_config* avgpool_config; const struct xnn_gavgpool_config* gavgpool_config; const struct xnn_pavgpool_config* pavgpool_config; const struct xnn_reduce_config* reduce_config; }; const struct xnn_gavgpool_cw_config* gavgpool_cw_config; const struct xnn_ibilinear_chw_config* ibilinear_chw_config; const struct xnn_ibilinear_config* ibilinear_config; struct { const struct xnn_rmax_config* rmax_config; union { // For QU8. const struct xnn_lut32norm_config* lut32norm_config; // For F16 and F32. struct { const struct xnn_raddstoreexpminusmax_config* raddstoreexpminusmax_config; const struct xnn_binary_elementwise_config* vmul_config; }; }; }; // For softmax operator. const struct xnn_maxpool_config* maxpool_config; const struct xnn_prelu_config* prelu_config; const struct xnn_unpool_config* unpool_config; const struct xnn_zip_config* zip_config; struct { const struct xnn_xx_fill_config* fill_config; const struct xnn_xx_pad_config* pad_config; }; // For constant pad operator. const struct xnn_x8_lut_config* lut_config; const struct xnn_cmul_config* cmul_config; const struct xnn_unary_elementwise_config* copy_config; const struct xnn_transpose_config* transpose_config; struct { const struct xnn_reduce_config* rminmax_config; const struct xnn_unary_elementwise_config* convert_config; }; // For F32 to QD8 convert operator. }; struct compute_parameters compute[XNN_MAX_COMPUTE_INVOCATIONS]; union { struct argmax_pooling_context argmax_pooling; struct average_pooling_context average_pooling; struct channel_shuffle_context channel_shuffle; struct conv2d_context conv2d; struct dwconv2d_context dwconv2d; struct dwconv_context dwconv; struct elementwise_binary_context elementwise_binary; // PACKW GEMM GOI + GEMM are used together in Dynamic Fully Connected. struct { struct gemm_context gemm; struct packw_gemm_goi_context packw_gemm_goi; }; struct global_average_pooling_nwc_context global_average_pooling_nwc; struct global_average_pooling_ncw_context global_average_pooling_ncw; struct igemm_context igemm; struct lut_contiguous_context lut_contiguous; struct lut_strided_context lut_strided; struct max_pooling_context max_pooling; struct pad_context pad; struct pixelwise_average_pooling_context pixelwise_average_pooling; struct prelu_context prelu; struct reduce_context reduce; struct resize_bilinear_context resize_bilinear; struct resize_bilinear_chw_context resize_bilinear_chw; struct slice_context slice; struct spmm_context spmm; struct subconv_context subconv; struct subgemm_context subgemm; struct transpose_context transpose; struct floating_point_softmax_context floating_point_softmax; struct u8_softmax_context u8_softmax; struct f32_qd8_convert_context f32_qd8_convert; struct univector_contiguous_context univector_contiguous; struct univector_strided_context univector_strided; struct unpooling_context unpooling; struct vmulcaddc_context vmulcaddc; struct rope_context rope; } context; struct xnn_code_cache* code_cache; struct xnn_weights_cache* weights_cache; enum xnn_run_state state; }; XNN_INTERNAL enum xnn_status xnn_run_operator_with_index( xnn_operator_t op, size_t opdata_index, size_t operator_object_index, pthreadpool_t threadpool);
13,461
31.995098
122
h